Organizing Immigrants: The Challenge for Unions in Contemporary California 9781501728839

Recruiting the growing numbers of immigrants into union ranks is imperative for the besieged U.S. labor movement. Nowher

128 30 31MB

English Pages 272 [271] Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Organizing Immigrants: The Challenge for Unions in Contemporary California
 9781501728839

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

ORGANIZING IMMIGRANTS

Org~nizing

lmm1grants The Challenge for Unions in Contemporary California Edited by

Ruth Milkman

ILRPRESS an imprint of

CORNELL UNIVERSI1Y PRESS Ithaca and London

Copyright© 2000 by Cornell University All rights reserved. Except for brief quotations in a review, this book, or parts thereof, must not be reproduced in any form without permission in writing from the publisher. For information, address Cornell University Press, Sage House, 512 East State Street, Ithaca, New York 14850. First published 2000 by Cornell University Press First printing, Cornell Paperbacks, 2000 Printed in the United States of America Library of Congress Cataloging-in-Publication Data

Organizing immigrants: the challenge for unions in contemporary California I edited by Ruth Milkman. p. em. Includes bibliographical references and index. ISBN 0-8014-3697-4 (cloth)-ISBN 0-8014-8617-3 (pbk.) 1. Alien labor-Labor unions-Organizing-California. I. Title: Challenge for unions in contemporary California. II. Milkman, Ruth, 1954HD6490.072 U657 2000 331.6'2-dc21 99-050221

Cornell University Press strives to use environmentally responsible suppliers and materials to the fullest extent possible in the publishing of its books. Such materials include vegetable-based, low-VOC inks and acid-free papers that are recycled, totally chlorine-free, or partly composed ofnonwood fibers. Books that bear the logo of the FSC (Forest Stewardship Council) use paper taken from forests that have been inspected and certified as meeting the highest standards for environmental and social responsibility. For further information, visit our website at www.cornellpress.cornell.edu. Cloth printing Paperback printing

10

9 8 7 6 5 4 3 2 10

9

8

7

6

5

4

3

2

®~

Contents

Acknowledgments Introduction

Vll

1

Ruth Milkman

Chapter l

Who Does What? California's Emerging Plural Labor Force

25

David Lopez and Cynthia Feliciano

Chapter 2

Immigrant Workers and American Labor: Challenge ... or Disaster?

49

Roger Waldinger and Claudia Der-Martirosian

Chapter 3

"Organize or Die": Labor's New Tactics and Immigrant Workers

81

Rachel Sherman and Kim Voss

Chapter 4

Immigration and Unionization in the San Francisco Hotel Industry

109

Miriam J Wells

Chapter 5

Intense Challenges, Tentative Possibilities: Organizing Immigrant Garment Workers in Los Angeles

130

Edna Bonacich v

CONTENTS

vi

Chapter 6

Organizing Latino Workers in the Los Angeles Manufacturing Sector: The Case of American Racing Equipment

150

Carol Zabin

Chapter 7

Organizing the Wicked City: The 1992 Southern California Drywall Strike

169

Ruth Milkman and Kent Wong

Chapter 8

Union Representation of Immigrant Janitors in Southern California: Economic and Legal Challenges

199

Catherine L. Fisk, Daniel J B. Mitchell, and Christopher L. Erickson

Chapter 9

The Los Angeles Manufacturing Action Project: An Opportunity Squandered?

225

Hector L. Delgado

References

239

Notes on Contributors

249

Index

253

Acknowledgments

T

his volume had its origins in a conference held at UCLA in May 1998, which I co-organized with Roger Waldinger, then Director of the UCLA Lewis Center for Regional Policy Studies. My heartfelt thanks to Roger for his ongoing support and practical advice, which helped transform a vague idea floated in a casual e-mail into a palpable reality. The other person who contributed immeasurably to this project is Kent Wong, director of the UCLA Labor Center, who offered innumerable suggestions and helped involve the local labor community in the conference, greatly enriching both the discussions held there and ultimately the contents of this volume. In addition to the seed money provided by the Lewis Center, I am grateful to the many conference co-sponsors: the California Policy Seminar, UC Mexus, the UCLA Center for Social Theory and Comparative History, the UCLA Division of Social Sciences, the UCLA Associate Vice Chancellor for Academic Mfairs, and the Office of the President of the University of California. The administrative support of MargaretJohnson, Belinda Vigil, and Luis Escala Rabadan was also essential to making the conference a success. Thanks also to David Gutierrez, Sanford Jacoby, Peter Olney, and Paul Johnston, who served as commentators at the conference, stimulating valuable discussions and offering insightful comments on earlier versions of the chapters appearing in this volume. Fran Benson at Cornell University Press was enthusiastic about this book from the outset and offered consistent encouragement at every step in the publication process. Dan Clawson and Elissa McBride provided

vii

viii

ACKNOWLEDGMENTS

supportive and incisive comments on the conference proceedings that were extremely helpful in shaping the final version of the book. Thanks also to Laura Robinson for expert technical assistance in the final stages of preparing the manuscript for publication. RuTH MILKMAN

Los Angeles, California

ORGANIZING IMMIGRANTS

Introduction

Ruth Milkman

A

t the end of the twentieth century, the challenge of recruiting immigrant workers into union ranks has become increasingly central to the larger project of rebuilding the United States labor movement, which has been in a downward spiral for decades. Today only about one in ten private sector workers, and only 15 percent of the workforce as a whole, are union members-less than half the level of union density forty years ago. When legal changes sparked a resurgence of mass immigration to the United States after 1965, union membership levels and influence were already waning; and in the decades since, labor movement decline has accelerated even as the immigrant population has swelled. Today, recent immigrants are even less likely than native-born workers to be unionized. Yet in some regions, among them the major metropolitan areas of California, immigrants today comprise a majority of those employed in manual jobs with limited educational requirements-the very jobs that were once key strongholds of union organization. Thus recruiting immigrants is an increasingly urgent imperative for the besieged labor movement. There are some hopeful portents on the horizon. In the 1990s, a series of dramatic immigrant organizing successes-many of them documented in these pages-demonstrated the potential for bringing foreignborn workers into the orbit of organized labor. At the same time, a new group of progressive leaders has taken command of the American Federation of Labor-Congress of Industrial Organizations (AFL-CIO) and I am grateful to Michael Burawoy, Paul Johnston, Roger Waldinger and especially Dorothy Sue Cobble for their critical comments on an earlier version of this introduction.

2

RUTH MILKMAN

some of the major unions affiliated with it have begun pouring resources into organizing on a scale not seen for decades. Immigrant workers are a crucial focus in these new efforts, especially in California, the nation's single largest magnet for immigrants and a key arena in the struggle for labor movement revitalization. Are immigrant workers fundamentally different from their nativeborn counterparts when it comes to union organizing? Does the fact that most immigrants who come to the United States today are from Latin America and Asia make their propensity to unionize (or lack of it) different from that of the earlier wave of European immigrants in the late nineteenth and early twentieth century? Are temporary sojourners more or less "organizable" than long-term residents who have no intention of returning to their home countries; more generally, are there differences between the orientations of settled immigrants and newcomers? How does the legal status of immigrants affect the situation-do naturalized citizens, legal residents, and undocumented immigrants differ in regard to their interest in unionization? Once organized, do foreign-born workers have different needs than their native-born counterparts, shaping a distinctive pattern of consolidating unionism, once it is established? How much does the political and cultural background of immigrants matterare politicized Central American refugees, say, more likely to organize or be recruited into unions than campesinos from rural Mexico or impoverished migrants from Thailand? And how do all these variations among workers play out in the context of the dramatic structural transformations that have been underway since the mid-1970s? Is there any relationship between immigration and the bold anti-union offensives on the part of employers and the state in this period? What about the rapid growth of inequality in income and wealth that has accompanied economic restructuring? How does the newly intensified polarization between rich and poor affect foreign-born workers and the prospects for bringing them into the labor movement? In short, what are the social, political, and economic conditions that facilitate, and those that impede, immigrant unionization? These are among the questions the contributors to this volume explore, drawing on newly collected evidence on immigrants and unions in late twentieth-century California. The chapters include broad overviews of immigrant employment and unionization patterns in the nation's most populous state, as well as studies of particular organizing campaigns, strikes, and immigrant-employing industries in a variety of urban settings, with cases from both northern and southern California. Most of the em-

Introduction

3

phasis is on Latinos, who constitute the bulk of the state's working-class immigrants, although some chapters explore the situation of Asian immigrants as well. There have been few previous efforts to document and analyze the relationship of immigrants to unionism in the contemporary United States, but the basic issues involved here are hardly new. Indeed, the historical literature on immigrants and the labor movement offers some useful insights that can help make sense of the current scene. For the American working class always has been comprised overwhelmingly of immigrants and their offspring. From the outset, there were many instances in which "ethnic bonds ... served to cement labor organization" (Brody 1993, 108). On the other hand, the diversity of nationalities within the United States was a challenge for union organizers from the earliest period. In order to forge durable labor organizations, "a multi-ethnic workingclass subculture of opposition had to be created," historian Richard Oestreicher (1986) notes in one influential formulation of the problem. "New institutions had to be created, and a new moral code developed that convinced workers that they owed the same kinds ofloyalties to workers of other nationalities that they owed to people of their own nationality" (61). This task was never easy, and in the unions affiliated with the American Federation of Labor (AFL), which by the end of the nineteenth century was the nation's hegemonic labor organization, all too often it was not even attempted. Many of the craft unionists of northern and western European origins that dominated the AFL were openly hostile toward more recent immigrants from southern and eastern Europe, whom employers freely exploited as strikebreakers and who dominated the ranks of the unskilled. Although in the 1880s and 1890s the AFL had made some serious efforts to incorporate these "new immigrants" into their organizations, by the turn of the century most trade union leaders had come to see them as "unorganizable" and to favor restrictions on new immigration (Greene 1998). The most extreme exclusionary impulses were directed at non-European migrants-the case of Chinese workers in California is especially infamous-and pervaded even the most progressive unions and labor organizations (Mink 1986). In some AFL affiliates, the "new immigrants" organized themselves or were recruited into de facto industrial unions, but this effort produced lasting results in only a few industries-most importantly, clothing and coal mining. Otherwise, except in such short-lived organizations as the Industrial Workers of the World, they remained unorganized until the mass industrial-union drives of the 1930s.

4

RUTH MILKMAN

The problem was not lack of interest in unionization on the part of the "new immigrants" themselves, however. When the opportunity arose to organize, they responded with marked enthusiasm. Economist Issac A. Hourwich wrote in 1912 that "immigrants from Southern and Eastern Europe are the backbone of some of the strongest labor unions" (34). In the unionization drive in the steel industry during World War I, "the great initial response came from the immigrant steelworkers. They were the first to crowd the mass meetings and sign up for membership," David Brody notes (1960, p. 223). "In contrast, the natives were, according to all reports, an unenthusiastic lot." In meatpacking, too, "organizers found the foreign"born much easier to organize than the native-born" (Barrett 1987, 195). Similarly, in the needle trades, especially among left-wing east European Jews, "it was within the immigrant working-class community that the roots oflabor militance took hold and spread" (Glenn 1990, 178). Both for the largely Slavic immigrants in steel and meatpacking and for the Jewish garment workers-as for English and German immigrants half a century earlier (see Brody 1993, llO-ll)-receptivity to unionization efforts was often linked to prior experiences with strikes and labor organization in Europe (Hourwich 1912, 349-51; Glenn 1990, 178-85). Of course, not all "new immigrants" had this kind of background, and ethnic parochialism and fragmentation among workers could be an obstacle to unionization in some contexts. But on the whole, what held back immigrant labor organizing in the early twentieth century was not antipathy toward unionism among the foreign-born themselves. Rather, the key difficulties were intensive, state-supported employer resistance on the one side, and on the other, the structural limitations of the craft-dominated AFL, for whose leaders the idea of recruiting within the ranks of the unskilled, where the new immigrants were so highly concentrated, was anathema. Only with the triumph of industrial unionism in the 1930s were these workers fully welcomed into the house of labor, along with the longmarginalized Mrican American working class. As Lizabeth Cohen has argued, the homogenizing effects of mass popular culture in the 1920s, along with the highly deliberate efforts a decade later by the Congress of Industrial Organizations (CIO) to build a "culture of unity" that preempted ethnic fragmentation, yielded powerful results when economic and political conditions shifted in a direction favorable to union-building in the New Deal era. Now union leaders aimed "to meet workers on their ethnic, or racial, ground and pull them into a self-consciously common culture that transcended those distinctions" (1990, 339), and unskilled first- and second-generation immigrants from southern and eastern

Introduction

5

Europe became the core of the newly revitalized labor movement. Whereas earlier unionizing efforts among this population, even when successful, had failed to build lasting organizations, the new industrial unions (both those in the CIO and the many AFL unions that transformed themselves into industrial organizations in this period) soon were consolidated into vast bureaucratic empires that would endure for decades. As such they offered opportunities to those first- and secondgeneration immigrants with the requisite leadership skills to become career union officials; at the same time the CIO's institutional success brought rank-and-file foreign-born workers squarely into the economic mainstream, often for the first time. Paradoxically, then, the putatively oppositional industrial union movement was a vehicle of immigrant assimilation. It is probably significant in this regard that the bir:th of the CIO came a full decade after the flow of immigrants from Europe was cut off by legal restrictions on entry to the United States, when the foreign-born population had largely stabilized. The fragmentary evidence available indicates that secondgeneration immigrants predominated among CIO activists (Cohen 1990, 324-25; see also Friedlander 1975). Indeed, Michael Piore has suggested that the unionism of the 1930s "can be understood in large measure as a part of the process through which ethnic communities coalesced and the second-generation communities expressed their resentment against the job characteristics that the parental communities, with a different motivation and a different attitude toward the labor market, had come to accept" (1979, 156-57). Cultural assimilation went ,hand in hand with economic integration. Not only did their entrance into industrial unions bring economic stability and a measure of dignity for those first- and second-generation immigrants who directly shared in the material gains the CIO extracted on behalf of their members. But also, the CIO's success helped transform the nation's overall economic structure in a more egalitarian direction, raising wages (and later, extracting fringe benefits) for blue-collar workersamong whom immigrants and their children were disproportionately represented-who had previously eked out a marginal existence. Levels of inequality in income and wealth stabilized in this period, and from the end of World War II through the early 1970s, with U.S. hegemony at its height, living standards for the working class as a whole rose dramatically while poverty levels fell to their lowest point ever (Levy 1988). Although industrial unionism was hardly the only factor shaping this process, it played an important role (Freeman 1993).

6

RUTH

MILKMAN

In California and the Southwest (and on a smaller scale, in Chicago as well), the CIO helped further assimilation and Americanization among Mexican Americans in much the same way as it did for those from southern and eastern Europe. There had been extensive immigration from Mexico in the 1910s and 1920s, and Mexican Americans were by far the largest Latino immigrant group (Portes and Bach 1985, 76-83). Unionism now brought large numbers of Mexicanos, both those born in the United States and immigrant newcomers, substantial improvements in economic status. A number of Mexican American union leaders emerged in this period as well. Even when it had a distinctively left-wing character, as was the case more often than not, labor activism among Mexican Americans in this period "involved at its core an attempt by the children of the immigrant generation and those who had arrived in the United States as youngsters to integrate themselves into American society," historian George Sanchez (1993, 249) writes in his study of Los Angeles. "Ironically, such [left-wing] labor and political activity often served as the greatest 'Americanizing agent' of the 1930s and 1940s." But there were limits to Americanization in this case. As the southern and eastern Europeans who dominated the early twentieth-century foreign-born population were incorporated into the economic mainstream with the help of industrial unionism, they also became "white," shedding their previously racialized status. By contrast, like African Americans, Mexican Americans and other Latinos not only continued to be classified racially (despite the wide variation in their actual skin color) and subjected to systematic discrimination in many public settings, but they also remained economically disadvantaged relative to non-Hispanic whites. Still, in the postwar boom years, thanks in large part to the direct and indirect effects of large-scale unionization, Mexican Americans made substantial gains in income, occupational status, and educational attainment, narrowing the gap with the white majority (Gutierrez 1998, 312). The public image of Mexican American unionism in the postwar era is strongly associated with the United Farm Workers (UFW), thanks to its high-profile boycott efforts and the stature of its key leader, Cesar Chavez, in the 1960s. However, as had been the case with previous efforts to build agricultural unionism, the UFW's collective bargaining achievements proved fleeting (see Martin 1996). The UFW nonetheless became a key symbol of the potential of unionism for the Mexican American community, especially in California, and provided a crucible of organizing experience for Latino labor activists that would be an important resource in later unionization efforts. Although they never achieved the UFW's na-

Introduction

7

tional visibility, in the 1960s Mexican Americans also engaged in labor organizing efforts in urban settings, where settled immigrants were more likely to be employed (in contrast, farmworkers were disproportionately newcomers). Both the UFW and other union-building efforts attracted the energies of a new generation of Latino political activists influenced by the Civil Rights movement and the New Left-a group that included the college-educated children of blue-collar workers who had been unionized in the CIO era (Gutierrez 1995, chap. 6). Just as this wave of ethnic political activism was emerging among Mexican Americans, the 1965 amendments to the nation's immigration law set the stage for a massive influx of newcomers that would soon enlarge and transform the Latino community. Immigrants poured in from Mexico, and later, Central America as well, and many came without the benefit of legal documents. Most were poor and had little education-in contrast to the more diverse and often more prosperous population of Asian immigrants who were arriving at the same time. California, especially southern California, was the destination of choice for the new immigrants, and this phenomenon was amplified by patterns of chain migration. Most of the Latino newcomers clustered at the bottom of the state's labor market, in low-wage manual jobs with limited educational requirements. Whereas the postwar boom had opened up opportunities for immigrants and their descendants to improve their lot, the post-1965 immigrants entered an economy where such opportunities were becoming more constricted. Deunionization and economic restructuring combined to create an increasingly polarized social structure, with the masses of working-class immigrants concentrated in low-level blue-collar factory and service jobs as well as a wide variety of positions in the informal sector where subminimum wages, homework, child labor, and other illegal practices are widespread. Although for many, even the bottom of the labor market in the United States represented a vast economic improvement over what they had left behind at home, future prospects for these immigrants and their children may be quite different from those of earlier cohorts. As David Gutierrez (1998, 321) has written, "whereas ethnic Mexicans and their children in most areas in the postwar years had a reasonable chance of moving from entry-level, low status, unskilled and semiskilled occupations into higher-paying, higher status, skilled bluecollar occupations, in recent years Mexican immigrants have tended to enter the ... bottom of the economy-and to stay there." The same is true for other working-class Latino immigrants (and some Asian groups as well), although for most of them there is no settled comparison group of the same nationality.

8

RUTH MILKMAN

Could unionism once again serve as the vehicle for lifting the new immigrants out of poverty and into the economic and cultural mainstream, as it did for the last wave of mass immigration half a century ago? One issue that differentiates the current situation from that of earlier periods is the presence of large numbers of illegal, undocumented immigrantsa category which simply did not exist before the imposition of restrictions on immigration in 1924. Because of their vulnerability to deportation, one might expect the undocumented to be more fearful about the risks involved in union organizing, particularly when confrontations with state authority are likely. Yet this may be less of an obstacle to union mobilization than is generally presumed. "Undocumented workers' fear of the 'migra' [U.S. Immigration and Naturalization Service] did not make them more difficult to organize than native workers or immigrant workers with papers employed in the same industries," Hector Delgado (1993, 61) concluded from his pioneering study of a factory unionizing drive involving immigrants, most of them undocumented, in Los Angeles during the mid-1980s. Workers reported giving little thought to their citizenship status and the possibility of an INS raid of the plant. ... A forklift driver at Camagua [pseudonym for the company] claimed that he had never been afraid of the INS, adding, "''ve never seen them here. Only in Tijuana." ... [Another worker] said that he had a better chance of "getting hit by a car"-and he didn't worry about either.... In response to the prospect of deportation, ... workers responded that if deported they would have simply returned (in some cases, "after a short vacation"). Julia Real [pseudonym], a sewer, commented, "They're not going to kill you! The worse [sic] thing they [the INS] can do is send me home, and I'll come back." (61, 63) To be sure, recent efforts to tighten restrictions on immigration and renewed initiatives to deport the undocumented may have altered the climate in the years since Delgado did his fieldwork. Yet, despite the large numbers of undocumented immigrants among them, there is survey evidence, albeit fragmentary, suggesting that foreign-born workers' -andespecially Latinos'-attitudes are generally more favorable toward unions than are those of native-born workers, even though immigrants often have difficulty securing work in the unionized sector of the economy (See DeFreitas 1993). In the 1998 referendum on the anti-union Proposition 226 in California, which would have required unions to obtain members' permission to spend any of their dues on politics, 75 percent of the heavily immigrant Latino voting population opposed the measure, compared to only 53.5 percent of voters overall (Bailey and Shogan 1998).

Introduction

9

One reason for their more pro-union views may be that many of today's immigrant Latinos-especially those from Central Americahave some positive experiences of unionism in their home countries. Although this transnational influence is less extensive than it was for their early twentieth-century European predecessors, it is striking that many of the new rank-and-file immigrant union leaders have a history of union activism or left-wing political ties in their native lands. While many immigrant workers are from rural backgrounds, a substantial number arrive in the United States far more acquainted with the idioms of unionism and class politics than their native-born counterparts. Among the workers involved in the Justice for Janitors campaign in Los Angeles, for example, organizers reported a "high level of class consciousness," as well as a willingness to take the risks involved in organizing that was palpably shaped by experiences back home. "There, if you were in a union, they killed you," one organizer explained in discussing the Salvadorans' role in this effort. "Here ... you lost a job ar $4.25" (Waldinger et al. 1998, 117). The shared ordeal of immigration itself and the persistently high degree of stigmatization that foreign-born workers are forced to endure in their new home may also enhance the appeal of union organization. And, crucially, the strong social networks among immigrants-in marked contrast to the atomized communities typical of the native-born workforce in contemporary California-also facilitate organizing, as the Los Angeles janitors as well as other cases illustrate (see Delgado 1993; Milkman and Wong, this volume). Interest in unionism seems to be strongest among the growing number of immigrants who have been in the United States longer, and who plan to remain here (precisely those who voted against Proposition 226), whereas newcomers presumably have a different orientation. Immigrant workers, undocumented or not, seem ripe for organizing. The major impediment is not a lack of interest in unions on their part, but rather the intensely anti-labor environment which makes organizing workers of any type difficult in the late twentieth century. National Labor Relations Board (NLRB) representation elections-once the primary means of recruiting new members- had become a cruel charade by the 1980s, as employers learned to manipulate the system to subvert even the minority of elections that unions were able to win. Unfair labor practice complaints filed against employers with the NLRB skyrocketed in the 1970s .and 1980s (Freeman 1988). Among other things, illegal firings of pro-union workers are now a routine practice, as workers themselves are only too aware. "The fear of losing jobs, not the fear of being apprehended and deported by the 'migra,' was [workers'] principal concern"

10

RUTH MILKMAN

in the union organizing drive that Delgado studied (1993, 57). Employers have increasingly petitioned to decertify existing unions by means of NLRB elections as well. Even in workplaces where unions have been able to maintain a presence, there was a massive power shift from labor to management in the 1980s. Unions in one industry after another made dramatic concessions and "givebacks," eroding pay rates, benefits, job security, and what little worker control over the labor process they had managed to obtain in earlier years. Emboldened employers went on the offensive on every front starting in the late 1970s, and even where they failed to avoid or eliminate unions entirely, they drastically reduced labor's legitimacy and influence. In the political arena, unions found themselves increasingly marginalized and isolated from potential allies in other social movements, as public support for unionism plummeted. Meanwhile, workers' real wages stagnated, income inequalities soared to levels not seen since World War II, and employment insecurity became pandemic. The decline in organized labor's power and influence turned what had been a gradual process of erosion of union density into a rout, and by the 1980s union membership levels were in a free fall. The trend in California was similar to that in the United States as a whole: in 1961, over a third (34.5 percent) of all nonagricultural wage and salary workers in California were unionized; by 1987 the figure had dropped to only 19.1 percent (California Department oflndustrial Relations 1989, 3). Private sector union density had fallen to a postwar low of 10.0 percent in the United States by 1996, and to 10.4 percent in California (Hirsch and Macpherson 1997, 26). Although this precipitous decline was primarily due to the anti-union offensive launched by employers starting in the 1970s (Freeman 1988), it did not help matters that unions devoted minimal resources to new organizing efforts in the postwar period (Voos 1984). In the 1990s, as the crisis deepened to the point where its very survival was threatened, the labor movement finally began to reorient itself. The turning point was John]. Sweeney's October 1995 victory in the first-ever contested election for the AFL-CIO presidency, bringing a dedicated and savvy group of young, progressive unionists into key positions in the AFLCIO bureaucracy. Sweeney has made a strong commitment to diversifying the nation's labor leadership and has initiated a dynamic union revitalization program as well. He has devoted considerable financial resources to new organizing-30 percent of the overall AFL-CIO budget-and is encouraging unions to build on recent experiments with new organizing

Introduction

11

tactics that a few innovative union locals have undertaken (see Sherman and Voss, this volume). In California, more than anywhere else in the United States, meeting the larger challenge of rebuilding the labor movement inherently demands an effort to reach out to the vast and growing population of working-class immigrants. In many of the state's industries and occupations (including the janitorial and drywall cases analyzed in this volume), as wages and working conditions deteriorated after employers successfully eliminated or weakened unions, native workers simply exited and were subsequently replaced by immigrants. Thus even recapturing formerly unionized territory-much less making inroads into newer sectors that have never been organized, where foreign-born workers also have a huge presence-requires the labor movement in California to confront the issue of immigration head-on. There have been some promising developments on this score, especially in the state's southern region, where the influx of Latino immigrants into the labor force has been particularly dramatic, and where they suddenly emerged as a core source of union militancy in the early 1990s. Latinos have recently risen to prominent positions in the regional labor movement, most notably with the 1996 election of Miguel Contreras as the head of the Los Angeles County Federation of Labor, the first non-Anglo ever to hold that position (Sipchen 1997). And in February 1999, a stunning 74,000 Los Angeles County home-health-care workers, many of them Latino immigrants, voted to unionize after an eleven-year campaign by the Service Employees' International Union-the single largest organizing success anywhere in the United States since the 1930s (Greenhouse 1999). The future of the labor movement in California may depend on whether it can build on this groundwork to recruit the vast numbers oflow-wage immigrant workers in the state into its ranks. The following chapters explore the prospects for that occurrence from a variety of perspectives. We begin ""ith a broad overview of the role of foreign-born workers in California's economy, which has been disproportionately affected by recent immigration. Nearly one-fourth of the state's population is foreignborn, compared to less than one-tenth of the United States population as a whole. David Lopez and Cynthia Feliciano's analysis highlights the intense concentration of Latino immigrants, most ofwhom are from Mexico and Central America, in California's low-level manual jobs. Foreign-born Latinos are 17 percent of California's total workforce, but they constitute 36 percent of its service workers, 42 percent of its factory operatives, and fully 49 percent of its laborers. At the other extreme, im-

12

RUTH MILKMAN

migrant Latinos account for only 5 percent of the state's professional and technical workers. The situation of Asian immigrants, who make up 10 percent of the state's workforce, is quite different: they are much more evenly distributed through the occupational structure, thanks to higher levels of education and far greater diversity in class background. Lopez and Feliciano also examine employment patterns in the state's two most important metropolitan areas, greater Los Angeles and the San Francisco Bay Area. Latino immigrants constitute 19 percent of the workforce in the Los Angeles region, more than triple their share in the Bay Area (6 percent), where foreign-born Asians comprise a larger proportion of the workforce (11 percent). It is also striking that the relatively small Latino immigrant population in northern California has higher educational, occupational, and income levels than its southern counterparts-perhaps because unions have historically been stronger in the Bay Area than in Los Angeles. Lopez and Feliciano pose the question whether the state's Latino immigrants will, like past generations of European immigrants, ultimately move into the economic mainstream, or whether they are more likely to become a castelike group stuck at the bottom of the occupational hierarchy. To explore this vital issue, they assess recent data on the educational and early job market experiences of the emerging population of secondgeneration Latinos. The results are not especially encouraging. In contrast to their Asian counterparts, whose educational and employment levels are actually ahead of native whites', Latinos are lagging behind. "The Latino second generation will certainly experience social mobility in contrast to their parents, fr.om the bottom to the middle rungs," Lopez and Feliciano conclude, assuming that new immigrants continue to fill the bottom rungs. "But changes in the economy, plus the vastly greater size of the new second generation, make it difficult to predict that they will end up even as well off as today's Chicano working class." Arguably, the fate of these second-generation Latino immigrants hinges largely on the degree to which the current efforts to revive the labor movement in California succeed. The optimistic scenario would be a reprise of the CIO era, when a burgeoning union movement served as the vehicle for second-generation immigrants to move up economically and simultaneously to assimilate into the mainstream of American culture. In the second chapter of this volume, however, Roger Waldinger and Claudia Der-Martirosian present a more pessimistic outlook, based on their analysis of unionization patterns among immigrants and nativeborn workers in the United States as a whole and in California. They deploy Current Population Survey (CPS) data for this purpose, using a com-

Introduction

13

bined four-year sample from the mid-l990s. Such an analysis was not possible until recently, as the CPS began asking about respondents' place of birth only a few years ago. Waldinger and Der-Martirosian show that immigrants are underrepresented among union members nationally, and to an even greater extent in immigrant-rich California. They suggest several possible explanations for this. At the high end, for professionals and other highly educated middle-class immigrants, who constitute a greater proportion of the foreign-born today than in the early decades of this century, economic security and high incomes often can be achieved without union protection. At the other end of the class spectrum, unionization-especially if it involves short-term sacrifices-may have little appeal to lowwage "target earners" who are part of circular migration streams, even if they are the most exploited members of the foreign-born workforce, since their standard of comparison is to wages in their home countries, to which they plan to return. A key finding from this analysis is that recent immigrants (those arriving in 1990 or later) are the least likely to be unionized, whereas those who have been in the United States the longest (arriving before 1980) have unionization levels roughly double those of newcomers, and in California over four times as great. Although these cohort differences are reduced in size once controls for other variables are included in the analysis, they remain the most striking finding of this chapter. In fact, for the nation's most settled immigrants, union membership is as likely-and for most subgroups, more likely-as for native workers. In California, however, even foreign-born workers who arrived in the United States before 1980 are less likely to be unionized than natives, in contrast to the national pattern. Still, the cohort effect is powerful in California too, with a far greater gap between the unionization levels of natives and recent arrivals than between those of natives and long-settled immigrants. As Waldinger and Der-Martirosian recognize, it would be misleading to draw inferences from these findings about the attitudes or propensities of different groups in relation to unionism, for in the twentieth-century United States what determines whether a given worker is a union member is not his or her personal preferences but rather whether she or he is employed in a workplace where a union exists. Recent immigrants are often effectively excluded by language limitations or citizenship requirements from public sector employment-the one sector where unionization levels are high. Given this, together with the fact that nine out of ten private sector jobs are nonunion, it is hardly surprising that newcomers would be less likely than natives or more settled immigrants to find

14

RUTH MILKMAN

unionized jobs, which pay more than nonunion jobs and offer superior benefits. Waldinger and Der-Martirosian conclude on a pessimistic note that "immigration is unlikely to add a silver lining to the dark clouds facing American labor." Although they make a compelling case for the claim that recruiting immigrant newcomers into union ranks is a formidable task indeed, the dramatic differences they reveal among immigrant cohorts nonetheless may augur well for the longer-term prospects of unionizing the foreign-born population. If Lopez and Feliciano's predictions prove correct, that the new second generation will face serious obstacles to economic mobility, the prospects for recruiting them into the labor movement in future years may be enhanced still further. Whether the optimistic or pessimistic views about the labor movement's future ultimately prove to be correct may depend on the extent to which some of the successful new organizing tactics with which a few innovative unions have been experimenting in recent years are adopted more ·widely, and on whether they can be used to expand union recruitment of immigrant workers in particular. This is the focus of Rachel Sherman and Kim Voss's chapter, which draws on fieldwork in northern California to analyze the conditions under which local unions deploy such new tactics and to assess their effectiveness when applied to predominantly immigrant workforces. Their starting point is the observation that, despite evidence that the aggressive rank-and-file strategies recently adopted by a few leading-edge unions have proven highly effective, most unions have continued to use traditional methods. By comparing three groups oflocal unions, which they label "full innovators," "partial innovators," and "mandated innovators," they identify the conditions under which the new organizing strategies are most likely to be adopted. They also assess the impact of innovation on local unions' ability to organize immigrant workers. The focus on local unions here is salutary, since the success of the new AFL-CIO leadership's efforts to revitalize the labor movement ultimately will depend on what happens within its various affiliates, especially at the local level where organizing efforts are concentrated. Extending the revolution that has occurred at the top of the labor federation into these long-established structures is among the most difficult tasks that lie ahead. Thousands of mid-level union officers and staff members remain loyal to entrenched internal political machines, and many are hostile to Sweeney and his allies. As Sherman and Voss point out, even shifting from the familiar emphasis on "servicing" existing members to an ap-

Introduction

15

proach that puts new organizing at the center is experienced as threatening by many traditional union leaders. ' The locals that are "full innovators" commit extensive resources to new organizing and to campaigns that involve a high level of rank-and-file participation, a focus on issues of dignity and justice in the workplace, and aggressive, confrontational tactics vis-a-vis management, such as civil disobedience. These locals often pursue "corporate campaigns" that attack an employer's public image. Rather than simply going after "hot shops" where worker militancy has already erupted and where a desire for union representation is apparent-the only organizing which more traditional unions engage in at all-these locals use strategic targeting in choosing organizing sites. They also adopt various novel techniques to avoid the pitfalls of the traditional NLRB election process. Sherman and Voss identify three key conditions that, in combination, facilitate such innovative organizing: ( 1) a crisis of survival for the local union, usually precipitated by serious membership losses in its historical jurisdiction; (2) support from the International union with which the local is affiliated-often involving a top-down process of introducing tactical innovation rather than one emerging from the grass roots; and (3) the presence on the local union's staff of activists from other social movements, such as veterans of the 1960s, schooled in more militant and participatory mobilization tactics than those in labor's conventional arsenal. Foreign-born organizers on the staffs of these locals, similarly, often can draw on political experience in their home countries that facilitate the innovation process. Sherman and Voss argue that the full innovators are the locals most likely to succeed in organizing immigrants, partly because their leaders tend to be more sensitive to the needs of rank-and-file workers generally. V\Then they target immigrant-employing workplaces, these locals are more likely than others to hire bilingual organizers and to integrate issues of specific concern to immigrants into their campaigns. Yet the full innovators seldom seek to organize immigrants qua immigrants, that is, to the exclusion of native-born workers. These locals choose organizing targets for strategic reasons, and if those targets are found to employ significant numbers of immigrants, as many in California inevitably do, the full innovators' rank-and-file orientation leads them to respond more effectively than other locals to the specific needs of such workers and often to frame their campaigns around issues of immigrants' rights as well. In short, the full innovators are more effective than other locals at organizing generally, and this makes them more effective at organizing immigrants in particular.

16

RUTH MILKMAN

Complementing Sherman and Voss's overview of organizational and tactical innovation is Miriam Wells's case study of a single local union, Local 2 of the Hotel Employees and Restaurant Employees (HERE), which has long represented workers in the San Francisco hotel industry. This is an example of a unionized industry where the workforce has undergone dramatic shifts in composition, from a predominantly nativeborn population in 1970 to one in which immigrants (mostly Chinese, Filipino, and Salvadoran) made up 48 percent of the workers in 1990. In response to this transformation, Local 2 has successfully adapted, by effectively reaching out to its new foreign-born members. Although this case involves internal organizing rather than new recruiting, it does fit Sherman and Voss's model to some extent. The local union recruited onto its staff community activists and others with experience in social movements outside of organized labor, many of them bilingual, with support from the International HERE. This recruitment enabled the local union to preempt an internal crisis (Sherman and Voss's third condition for full innovators), although a major strike in 1980 did threaten to become just such an occasion. Wells analyzes the occupational distribution of immigrants within the San Francisco hotel industry, demonstrating that they are overrepresented in "back-of-the-house" jobs such as food preparation and room cleaning, although they have made some inroads into the industry's better jobs as well. She finds that most occupations, even within a single hotel, usually include workers of more than one nationality, so that "individuals with cultural differences also shared economic interests." Wells emphasizes that the great diversity within the local union's membership led it to avoid focusing on the needs of any one ethnic or national group, but rather to mobilize around the unifying experience of work. At the same time, Local 2 has actively sought to meet the specific needs of its foreign-born membership, including providing assistance with immigration-related problems. A major contribution of this chapter is its analysis, based on extensive interviews with union staffers, of the variations among immigrants of different nationalities in receptivity to the union's efforts. Wells suggests that Asians, with the exception of Filipinos, generally have a less positive view of unions than Latinos, especially Central Americans. She attributes these differences to workers' experiences (or lack thereof) with labor organizations in their home countries, and in particular to the "dangerous authority-challenging struggles" in which many foreign-born Latinos engaged in the course of the immigration process itself. Wells reports that despite the variation among immigrants, as a group they are generally

Introduction

17

more responsive to the union's efforts than native-born whites. (Mrican Americans, in contrast, are strongly pro-union, but they make up only a tiny part ofLocal2's membership today.) The geographic context for both Sherman and Voss's empirical analysis and Wells's case study is the San Francisco Bay Area, the part of California where unions have historically been strongest-although with the deunionization of the 1970s and 1980s, the regional gap has narrowed somewhat. In 1961, according to data collected by the state of California, 46 percent of nonagricultural wage and salary workers in the San Francisco-Oakland area were unionized; a figure that had dropped to 26 percent by 1987. In Los Angeles-Long Beach, by contrast, union density was always lower, declining from 32 to 20 percent over the same period (California Department of Industrial Relations 1963 and 1989). By 1996, according to CPS data, the gap between these two areas was even smaller: in the Los Angeles-Riverside-Orange county area, 15.4 percent of the workforce were union members (slightly higher than the national figure of 14.5 percent), while in the area including San Francisco, Oakland, and San jose, the figure was 18.9 percent-a figure that may be somewhat misleading since unions were never as strong in San Jose as in San Francisco and Oakland (Hirsch and Macpherson 1997,51, 55). San Francisco is distinctive in another important dimension: as Waldinger and DerMartirosian note (see table 2.4, this volume), the gap in unionizr.tion levels between immigrants and native-born workers is much smaller there than in the rest of the state. In southern California the dilemmas involved in immigrant unionization are especially stark. Not only have unions historically been weaker there than in the north, but also, as Lopez and Feliciano show, the immigrant influx into the workforce has been far more extensive than in the northern part of the state, particularly among Latinos. There have been numerous immigrant organizing efforts in southern California-some successful, some not- which are the subjects of the remaining chapters in this volume. An extreme example is the case of the garment industry, the largest surviving manufacturing industry in the region, which is the focus of Edna Bonacich's chapter. This industry, which accounts for nearly a fifth of all manufacturing employment in Los Angeles, represents the very bottom of the labor market for immigrants, attracting mostly female, often undocumented, recent arrivals who earn substandard wages and labor under notoriously poor conditions. Because garment-making is a prototypically global industry, with much of its production long since shifted offshore, downward pressures on labor costs are especially intense, and employers' resistance to unions

18

RUTH

MILKMAN

is fierce. The industry was once centered in New York, where the relatively few garment workers who remain are still highly unionized. But in Los Angeles, which today has more garment industry employment (over 120,000 workers) than any other part of the United States, only a few hundred of the industry's workers are unionized, despite ongoing efforts by the Union of Needletrades, Industrial and Textile Employees (UNITE) to recruit them. In view of the recent successful efforts to unionize immigrants in Los Angeles in other industries on the one hand, and the history of unionism in the garment industry (which has long relied heavily on immigrant labor) in other times and places, why has organizing proved so difficult? Bonacich points out that many of the recent immigrant organizing successes in southern California involve work that is not geographically mobile, such as janitorial services and construction. By contrast, garment factories threatened with unionization can readily shift production offshore. As for the poorly paid immigrant women from southern and eastern Europe who did manage to unionize in early twentieth-century New York, against great odds, Bonacich suggests that the sophistication of the manufacturers and the ease with which they can relocate production on a global scale is far greater today than it was then. She emphasizes that the problem is not a lack of interest in unionization among rank-and-file garment workers but rather their knowledge that organizing is likely to lead to job losses and that the chances of victory are slim. The traditional route to union representation, namely NLRB elections, is virtually irrelevant under current conditions in this industry, which relies heavily on subcontracting. Any factory in which UNITE wins an NLRB election is simply avoided by the manufacturers, and is therefore likely to go out of business, as has occurred many times already in Los Angeles. Faced with this grim reality, UNITE has recognized the need to target manufacturers rather than subcontractors, or even better, to try to organize an entire sector of the industry at once. The current corporate campaign against Guess, Inc. is an example of this type of targeting, although thus far the company has resisted the union's efforts fiercely. As Bonacich points out, such large-scale organizing drives involve a huge commitment of resources, which UNITE may be reluctant to make given the enormous obstacles to success. Yet organizing immigrants in the globalized manufacturing sector is not completely impossible, as Carol Zabin's chapter shows. She analyzes a wildcat strike that erupted in the summer of 1990 among the twdve hundred workers at the American Racing Equipment (ARE) wheel factory in Los Angeles, which led to union recognition the following year. Although

Introduction

19

the company does have a plant in Mexico, where a great deal of auto parts production has been moving in recent years, it also has some good reasons to continue production in Los Angeles, so that the threat of plant relocation or closure in response to unionization is less dire than in the apparel case. The workforce at this plant is quite different from that in the garment industry, even though both are comprised almost entirely of Latino immigrants: unlike the female newcomers who dominate the apparel workforce, most of the ARE workers ·are long-settled male immigrants-confirming Waldinger and Der-Martirosian's finding that immigrants of the longest vintage are more likely to be unionized. Zabin emphasizes the fact that the ARE strike and the union recognition campaign that it galvanized were traditional in character. Although some of the innovative tactics that Sherman and Voss discuss were employed, this was a prototypical "hot shop." A previous attempt to unionize the plant two years before had failed, but now rank-and-file workers, infuriated by technological and organizational changes, organized on their own. Their three-day wildcat strike came to the attention of the labor movement only after it was reported in the Los Angeles Times, at which point a number oflocal unions made overtures to the workers. The ensuing union drive took a traditional form, organized around the goals of winning an NLRB representation election and, subsequently, a first contract. The International Association of Machinists and Aerospace Workers (lAM), which emerged the victor in the interunion competition to take up the organizing opportunity that the wildcat strike presented, by all accounts conducted an exemplary campaign. It rented an office near the plant that was open twenty-four hours a day, assigned several bilingual Latino organizers to the effort, and developed a rank-and-file workers' committee which was integral to the NLRB election campaign. Tactics included house calls, barbeques, and generally promoting the idea that the workers themselves "owned" the campaign. Despite the fact that ARE hired an anti-union "labor consultant," the union won the election in a close vote in December 1990. Next came a nine-month campaign for a first contract, an effort which included extensive worker education and training and a variety of in-plant shows of force to maintain pressure on the company. The first contract brought fairly minimal benefits (as is often the case), but since that time the union has improved its terms significantly, and management seems to accept the union at this plant as a fact of life. Two of the most talented rank-and-file leaders who emerged in the ARE strike were hired as Machinists' Union staffers, although not

20

RUTH MILKMAN

for some years after the strike. Yet as Zabin emphasizes, the union failed to take the other logical follow-up step to its victory at ARE, namely to attempt to organize the rest of the wheel industry in the region. Among twenty-five wheel plants in greater Los Angeles, only two (including ARE) are unionized-and even ARE has a second, nonunion plant in the area. Despite talk of an industry-wide campaign, this has yet to be seriouslrattempted. Still, the ARE case suggests that immigrants in the manufacturing sector can, at least under certain conditions, be successfully unionized. Just as the union at ARE was securing its first contract, an even larger organizing effort was emerging among southern California's immigrant workers. As Ruth Milkman and Kent Wong's chapter recounts, thousands of Mexican drywall hangers, squeezed by wage cuts during the severe residential construction slump of the early 1990s, began a union organizing effort in late 1991, culminating the following year in a successful fivemonth strike that shut down housing construction in six southern California counties. As in the ARE strike two years earlier, the initiative in the drywallers' strike came from the grass roots. Workers began to organize entirely on their own, and indeed their effort met with considerable skepticism by the Carpenters' union that had represented residential drywallers prior to the trade's deunionization in the late 1970s and early 1980s. Although the Carpenters did allow the strikers to use their halls for meetings, there was no official union involvement until the strike's end, when the Carpenters negotiated a settlement and contract, under which 2,400 drywall hangers ultimately became union members. The strike was partly built on the basis of social networks among the immigrant workforce: most of the leaders and many others who participated in the campaign's early stages were from the same small village in Mexico. Once the strike was underway, the organized labor movement rose to the occasion and provided extensive financial support as well as legal assistance. There was enormous community support for the effort as well, with Latino immigrant organizations and church groups providing food and other forms of support. The California Immigrant Workers' Association (CIWA), sponsored by the AFL-CIO, played a crucial role also, mainly through its legal work. Yet all of these support efforts were stimulated by the rank-and-file's own mobilization efforts. Ultimately legal issues became central to the strike's success, and at the same time limited the scope of the victory. On the positive side of the ledger, what finally induced the contractors to sign an agreement settling the strike and recognizing the union was a series of CIWA-funded lawsuits alleging violations of the overtime provisions of the Fair Labor Standards

Introduction

21

Act (FLSA). Labor practices of dubious legality had become widespread in the residential drywall industry during the 1980s, after the union had been eliminated, and the employers proved extremely vulnerable to the FLSA suits, which were dropped in exchange for union recognition at the end of 1992. Meanwhile, however, a group of San Diego drywall employers, who refused to become part of the strike settlement, opened a second legal front to counterattack the Carpenters' union, filing suit against it under the Racketeer Influenced and Corrupt Organizations Act (RICO), which has increasingly been used against labor unions in recent years. This suit, which accused the union of orchestrating the strike and of responsibility for various acts of violence allegedly associated with it, was eventually settled out of court in the employers' favor for an undisclosed sum. The drywallers' strike, like the ARE case, demonstrates the potential for large-scale organizing among the Latino immigrants who increasingly dominate the manual labor force of southern California, even under the inhospitable conditions that unions face today. In both these cases success was predicated on the combination of bottom-up organizing among immigrant workers themselves, partly rooted in immigrant social networks, on the one hand, and the labor movement's willingness to commit extensive financial and legal resources to supporting the effort, on the other. Yet again, in both cases, organized labor failed to build on the momentum of these struggles. Just as the Machinists' union made no serious effort to organize the other wheel factories in the area after the ARE strike, so too there was hardly any follow-up on the drywallers' victory in terms of new organizing among the vast numbers of nonunion residential construction workers-many of them Latino immigrants as well-in other trades. Worse still, since the 1992 strike victory, the Carpenters' union has actually lost ground within the drywall industry, as nonunion contractors have taken over an increasing share of the market. The problem of union consolidation in the aftermath of organizing success is the focus of the chapter by Catherine Fisk, Daniel Mitchell, and Christopher Erickson. The janitors' Los Angeles organizing drive, which culminated in a strike in 1990 that led to a pathbreaking contract settlement, was initiated by the Service Employees' International Union (SEIU), which was headed at that time by current AFL-CIO President John Sweeney. In contrast to the ARE and drywaller strikes, it began as a top-down effort, although the campaign did involve extensive ground-level mobilization among immigrant janitors themselves as well. The story of the organizing, recapitulated briefly by Fisk, Mitchell, and Erickson, has been told in detail elsewhere (see Waldinger et al. 1998;

22

RUTH MILKMAN

Savage 1998). This chapter's main contribution is its analysis of the aftermath of the effort, examining the critical issue of sustaining the initial achievements of the organizing campaign and building a durable union structure. In the immediate aftermath of the 1990 victory, the SEIU local with which the janitors were affiliated became mired in factionalism, pitting those who wanted to put resources into continued organizing against others who wanted instead to concentrate resources on servicing the existing membership-a dilemma that many unions face after major organizing victories. The internal turmoil ultimately led the International union to put the local into trusteeship. In addition, the local union confronted challenges on a number of fronts. One involves maintaining the hard-won union share of the office building cleaning market, which is continually threatened by the possibility that union-cleaned buildings will revert to nonunion status and by the emergence of new nonunion buildings as the regional market expands. So far the union has held on to its market share (in contrast to the drywallers, who have already suffered serious decline in this respect), but the potential for erosion is ever present. As Fisk, Mitchell, and Erickson note, assuming economic growth continues in the region, protecting its existing membership ultimately will require the Los Angeles janitors' union to embark on aggressive new organizing just to maintain its current market share. This prospect, in turn, demands that the union confront a number of legal challenges that have emerged since the 1990 contract victory in Los Angeles. Recent court decisions have limited the kinds of tactics that SEIU-known for its creative and highly confrontational "in-your-face" style of militancy-will be permitted to employ in future janitorial organizing. Most important are concerns involving the delicate issue of longprohibited secondary boycotts (actions targeting not the immediate employer, which in the case of janitors is an office cleaning contractor, but other businesses such as building owners or tenants). Recent case law limits the union's right to engage in picketing as a way of mobilizing public sympathy and pressuring employers, as previous janitors' campaigns often did. Such issues have not become particularly salient in southern California thus far, simply because the janitors have done little in the way of new organizing since their 1990 victory, but they are potentially important concerns for the future. In 1990 the janitors won a union contract, as did the ARE workers the following year. Then came the 1992 drywallers' strike and victory. These struggles together brought several thousand of southern California's im-

Introduction

23

migrant workers into the union fold in rapid succession, creating a sense of momentum that many in the regional labor movement hoped would be the beginning of a massive wave of immigrant organizing. Yet since 1992 there have been few new breakthroughs, and none involving such large numbers of workers as these three key campaigns did. If these examples served to liquidate any lingering notions that immigrant workers are "unorganizable," they also indicate that enormous resources are needed to carry out such efforts. What they all shared was a dynamic rank-and-file mobilization effort and a major material and moral commitment from existing unions-in the janitors' and drywallers' cases amounting to millions of dollars. Even with such vast outlays, consolidating unionism in the aftermath of organizing success and continuing to recruit new members on the foundation these victories established has proven extremely problematic. In the moment of hopefulness that these three success stories produced, in the early 1990s a group of veteran Los Angeles-based labor organizers conceived of a new endeavor, which they called the Los Angeles Manufacturing Action Project (LAMAP). Hector Delgado's chapter, the last in this volume, analyzes the trajectory of this effort and the reasons for its ultimate abandonment only three years after it was officially established in 1994. The key LAMAP concept was strategically targeted, industry-wide organizing campaigns in Los Angeles' vast manufacturing sector, which employs about 700,000 workers, most of them Latino immigrants. The project included a substantial research component; with assistance from faculty members in the UCLA Urban Planning program as well as numerous students. From the outset, links to ethnic and community groups were also central to the project, which explicitly targeted immigrant workers. LAMAP envisioned multiunion campaigns, with financial contributions pooled from a wide array of labor organizations. The original notion was that any workers who were recruited through these efforts would become part of a new entity, rather than affiliating with individual unions, but this idea was soon abandoned in the face of resistance from some of the unions involved. As Delgado emphasizes, the traditional organizational culture of unionism in this respect proved incompatible with the more collective, movement-building vision LAMAP tried to promote. In any case, a necessary condition of the project's success was a serious financial commitment on the part of multiple unions. Though the project did receive seed money from a number of unions, when the time came to make a more serious financial commitment, almost all of them backed off.

24

RUTH

MILKMAN

Unfortunately, the timing of the effort coincided with the internal revolution within the AFL-CIO that culminated in the election of john Sweeney to the presidency in 1995. Despite the fact that LAMAP's ideas about organizing were almost perfectly congruent with those of the new leadership, the turmoil within the AFL-CIO's affiliates precipitated by Sweeney's ascent created new obstacles to bringing the project to fruition. That LAMAP had been conceived during the previous AFL-CIO administration, which even had provided it with some preliminary support, tainted it in many eyes as being associated with the "old guard." In addition, the project now could be seen as competition for the organizing efforts the Sweeney regime itself wanted to undertake. Ironically, these concerns ultimately led the AFL-CIO to decline to support it. In the end, only the International Brotherhood of Teamsters was willing to make a serious financial commitment, and though this did lead to some small-scale organizing efforts, it was insufficient to allow LAMAP's talented organizing staff to even attempt to carry out their original, highly ambitious plans. Not long after, the Teamsters themselves became embroiled in internal turmoil with the result that in 1997, even their support was cut off. Thus the most explicit effort to build on the lessons of the immigrant organizing successes in the early 1990s proved abortive. Nonetheless, LAMAP had an important impact on the thinking of labor movement activists in southern California, and to some extent nationwide, who absorbed both the vision of multiunion, citywide organizing that it embodied, as well as the lessons of its sad demise in the context of internal labor movement politics. One can only hope that if and when the new labor movement overcomes its growing pains, other attempts will emerge to pursue the immigrant organizing agenda defined by LAMAP and some of the other organizations whose efforts are documented in these pages. If this volume helps in that effort, even in a small way, it will have amply fulfilled its purpose.

l

Who Does What? California's Emerging Plural Labor Force David Lopez and Cynthia Feliciano

I

n the past twenty-five years immigrants have increased threefold as a proportion of California's workforce, and now constitute at least one-third of the total. Immigrant workers are among the most exploited of employees in California, they are the majority of workers in many settings that seem ripe for organizing, and they may carry with them a spirit of worker solidarity that has largely died out among U.S. workers. At the same time, they pose the traditional challenges to labor organization of any desperate and abundant labor force, and many face the special challenges of illegal status. This chapter provides an overview of the role that immigrants play in California's labor force. 1 We begin with a review of the remarkable demographic changes the state is experiencing, and the implications for the future. Then we trace the growth of the immigrant workforce and look at its ethnic composition in particular occupations and industries. Finally, we pose questions about the future economic role of the children of recent immigrants, the emerging second generation. The very phrase "immigrant workers" is ambiguous. About half of all immigrants in the workforce are manual workers falling into the laborer, 1. Our statistical data are drawn largely from two sources: the decennial U.S. Censuses and, for the most up-to-date information, the Census Bureau's Current Population Surveys done in March 1995, 1996, and 1997. Decennial Census data for 197Q-90 are derived from analysis of the Integrated Public Use Microdata Sample (IPUMS), the compilation of census data from 1850 through 1990 that eases the burden of making comparisons over time. Some census data come from published reports by the Bureau of the Census, as specified in each table. We have combined the nonduplicated respondents to the March 1995, 1996, and 1997 Current Population Surveys to provide one unified sample, which we refer to as the 1996 CPS.

25

26

D A V I D L 0 P E Z and C Y N T H I A

FE LI C I A N 0

operative, and service occupational categories, and these are the focus of our attention in this chapter. But conversely half hold white-collar or comparatively well-paying craft jobs. In fact, one of the distinctive characteristics of immigratiot1 today is its considerable socioeconomic diversity, particularly among immigrants from Asia. Indian physicians, Taiwanese engineers, and Mexican professors have arrived in the United States recently, along with Salvadoran peasants and Chinese factory workers. A comprehensive understanding of the role that immigrants play in California's economy must consider all of these, as well as the immigrant entrepreneurs and managers under whom immigrant workers often toil. But our focus here is on those at the lower end of the occupational spectrum: semiskilled and unskilled factory workers, laborers, and service workers, and we shall reserve the term "worker" for them. Such jobs constitute a small and declining proportion of the California job market, which is nearly two-thirds white-collar. There are almost four times more managerial, professional, and technical employees in the state than factory workers. But immigrants are disproportionately found in the humbler jobs, and they also make up larger and larger proportions of all manual workers in the state. The racial I ethnic dimension of immigration is fundamental to understanding California's changing demographics and especially the composition of the state's army of manual workers. 2 Over 70 percent of California's immigrant manual workers are Latino, overwhelmingly from Mexico and Central America. In many specific occupations immigrant Latinos now constitute one-third or more of all workers. Within thirty years immigrant Latino workers and their children will almost certainly comprise the majority of California's semi- and unskilled workers. This transformation has already occurred in the state's agricultural regions, and it is also nearly complete in the Los Angeles area. In the San Francisco Bay Area the Latino immigrant presence is less striking; there are nearly twice as many Asian immigrants in the labor force. Will ethnic concentration enhance worker solidarity and facilitate labor organizing? Or 2. Throughout this chapter our use of racial/ ethnic terms follows conventional usage in California, which corresponds to U.S. Census categories: Asian American, African American, white (the "non-Hispanic White" census category) and Latino ("Hispanic Origin"). The first two are technically "racial" tenns according to the Census Bureau, and the last is a unique sort of "ethnic" category, used only with reference to Latinos. Each term is based on a host of complex and contradictory "racial" and "ethnic" meanings and are umbrella terms covering a wide variety of subgroups. These four terms are also the most commonly used racial and ethnic labels and categories used in California, past and present, and are therefore the best to use in broad analyses such as that presented in this chapter. "Other" races amount to less than one percent of California's population.

Who Does What?

27

will its primary effect be to increase animosity and ethnic tensions? Whatever the result, we are certain that the salience of ethnicity will increase in the economy, as it will in all sectors of life in California. Like most newcomers to the United States throughout its history, immigrants and migrants from other states have usually been relegated to the bottom rungs of the occupational ladder in California. Most EuroAmerican and Asian American subgroups have reached rough parity within a generation or two, in contrast to Mrican Americans who continue to have markedly lower rates of upward mobility (Blau and Duncan, 1967). What about Latinos? Thirty years ago, in their massive study The Mexican-American People, Grebler, Moore, and Guzman (1970) argued that the future of Mexican Americans might go in either of two directions: they might proceed along a somewhat delayed path of assimilation and increased economic equity, like Italians and other Euro-American ethnic groups; or they might share the castelike fate of Mrican Americans. 3 The answer was to be found in the future, as third- and fourth-generation Mexican Americans became the majority of the group's population. What Grebler and his associates did not foresee was the great revival of immigration that has taken place in the past thirty-five years. Today Mexican and Central American immigrants and their young children have overwhelmed the state's older Mexican American population. This demographic transformation has greatly complicated the question posed by Grebler and his associates, and the immigration of poorly educated Latinos concentrated in low-paying manualjobs has contributed to the appearance of a castelike status for Latinos and growing ethnic stratification for the state as a whole. But even a minimal definition of "castelike" requires that low socioeconomic status be passed on from generation to generation. The social progress of third generation and subsequent Latinos in California and the United States generally is itself a complicated and contentious topic, but the balance of evidence suggests that they are not approaching parity with Euro-Americans. 4 Whatever the fate of these older generations, the "place" of Latinos in California society is increasingly determined by the large and growing population of immigrant workers and their children. There is little doubt that immigrants will con3. The term "castelike" here refers to groups or societies in which occupational and social status are strongly correlated with ethnicity, from one generation to the next. We use the term descriptively to characterize social outcomes. without implying that these outcomes are the result of political enforcement such as segregation or apartheid, or religious sanction as in South Asia. 4. Bean and Tienda ( 1986) is the most comprehensive analysis of generational differences among Latinos; Ortiz (1996) provides evidence that successive cohorts of U.S.-born Latinos in the Los Angeles region may be doing worse, not better.

28

D A V I D L 0 P E Z and C Y N T H I A

FE LI C I A N 0

tinue to do the state's dirty work. The crucial question is what kind ofjobs their children, the emerging second generation, will manage to attain. Will they gain the education necessary to move into the good jobs of tomorrow? Or will they be channeled into dead-end jobs only marginally more attractive than those taken by new immigrants? Or will the result be somewhere in between? Ethnic Change in California Driven largely by immigration since 1965, California is now the most multi-ethnic state in the country. Non-Hispanic whites ("Anglos") now constitute only about half the population. Mrican Americans continue to decline as a proportion of the state's population, with Asian Americans, at 12 percent of the total, now outnumbering them nearly two to one. Constituting nearly one-third of the state's population, Mexican Americans and other Latinos will be the single largest group early in the next century. California is undergoing a demographic transformation second only to the Yankee invasion 150 years ago, and unique in the recent history of the United States. The profundity of these changes becomes clear by comparing today's ethnic profile with California in the past. California has been multiethnic since the arrival of the Spanish, but for the past century and a half the state has been dominated demographically as well as politically by "Anglo" whites. Up to the middle of this century, California's labor force (like its population) was nearly 90 percent white. Asians were never more than 10 percent of the population, and that share declined to 2 percent in 1950. Mrican Americans constituted no more than 2 percent of the population until World War II brought the great migration from the South; their presence in California peaked at 7 percent in the postwar period and is now slowly declining. Despite the Mexican I Spanish origins of the state, it was only after the turmoil of the Mexican Revolution that the state's Latino population began to grow above the 3 percent level, to 10 percent in 1960. 5 As figures 1.1 and 1.2 show, all that began to change in the 1960s. By 1960 California had reached its low point in foreign-born population (though it always remained above the national average). Figure 1.1 shows the dramatic turnaround: the percentage of foreign-born has increased 5. We have drawn or estimated these figures from U.S. Census of Population: General Social and Economic Characteristics, for 1950, 1960, 1970, 1980, and 1990; Persons of Spanish Surname 1960, Persons of Spanish Surname 1970, Persons of Spanish Origin 1970. .

Who Does What?

29

dramatically since 1970, and the rate of increase shows no sign of abating. The radical nature of this shift is reinforced by comparing the sharp rise in California's immigrant presence with the much more moderate change in the entire country. By the year 2000 over one-quarter of California's population will be foreign-born, compared to 9 percent for the nation as a whole. 0 And even that 3:1 ratio understates the relative impact of immigration in California for two reasons. First, a larger proportion of California's immigrants are recent arrivals, not senior citizens, and, relatedly, these young immigrants have produced a second generation of children who are still largely of school age. Indeed, this alone explains many of the current tensions around immigration in California. Of course the newcomers are not just immigrants but nonwhite immigrants, which further contributes to potential tensions. The other pair of lines in figure 1.1 chart the growth of the aggregate nonwhite population in California and the nation. In 1950 California was nearly as white as the nation as a whole; today its proportion of nonwhites is twice the national average. Figure 1.2 charts the growth of the four principal racial/ ethnic groups in California from 1970 to 2000. All ethnic segments have grown in absolute numbers, but while the Mrican American and white populations have grown slowly, the Latino and Asian populations have tripled in size in the past thirty years. The Asian growth rate has been faster, but it is Latinos who will displace whites as the state's single largest ethnic group, sometime around 2030 (California Department of Finance 1996). Indeed, the Latino presence is already comparable to that of whites in the state's public schools, each constituting about 40 percent of the enrollment overall. Latinos and Asians constitute ever-increasing proportions as one moves down grade levels (Linguistic Minority Research Institute 1996). Latinos and Asians are not just the two fastest-growing groups, they are also the youngest, guaranteeing that natural increase as well as immigration will be an important factor in the changing ethnic balance. Figure 1.3 highlights the importance of natural increase, the demographer's euphemism for babies that will need health care, schools, and other institutions. In fact, the single largest component of population growth from 1990 to 1995 was natural increase of the Latino population, not immigration. This is a young population, and even if their birth rates were not above average (they are, though not as much as alarmists think), 6. National Year 2000 projections come from the Census Bureau web site; California projections come from the state Department of Finance, supplemented by estimations from the 1997 Current Population Survey, which yield similar results.

D A V I D l 0 P E Z and C Y N T H I A F E ll C I A N 0

30 60 50

40 'E Q)

e Q)

a..

30 20 10 0

1950

1960

D Calif. foreign-born

1970

o u.s. foreign-born

1980

1990

•Calif. nonwhite

2000 • U.S. nonwhite

Sources: U.S. Census of Population: General Social and Economic Characteristics, 1950, 1960, 1970, 1980, and 1990; Persons ofSpanish Surname 1960, Persons ofSpanish Surname 1970, Persons of Spanish Origin 1970. California 2000 projections from California Department of Finance, California Demographics (Winter 1998). U.S. 2000 projections from Census Bureau web site.

Figure 1.1. Percentage foreign-born and nonwhite in California and U.S.

populations, 1950-2000. their low mortality rates due to the small portion of elderly and high birth rates due to the large proportion of women in child-bearing years would be enough to explain this spectacular rate of natural increase, which accounts for over half of all the population increase in the state from 1990 to 1995. Figure 1.3 provides information about immigration itself that may surprise some readers. According to state estimates, Asian immigration outpaced Latino immigration about two to one in the first half of this decade. These data may underestimate Latino immigration, however, largely because they take insufficient account of undocumented immigration. But even if Latino immigration is double the state estimate, Latino natural increase would still be more than twice as important, and still the single largest source of population growth in California. Of course this natural increase is not unrelated to immigration: young immigrant families and, increasingly, the maturing second generation (the children of immigrants) are disproportionately represented among

80

70 60

c

50

~ 40

Q)

0..30

20 10 0

1970 •White

1980 D latino

1990 • Asian

DBiack

2000 • Other

Sources: 1970-1990 U.S. Census of Population, Integrated Public Use Sample (IPUMS-95), Version 1.0. Year 2000 estimates based on 1997 Current Population Survey and estimates from California Department of Finance, California Demographics (Winter 1998).

Figure 1.2. Ethnicity of California population, 1970-2000.

1,400 1,200 1,000 800 600 400 200 0 -200 -400

White

Latino Net migration

Asian

Black

• Natural increase

Source: Revised estimate by California Department of Finance, California Demographics (Winter 1998).

Figure 1.3. Components of California population growth by ethnicity, 1990 to 1995 (in thousands) .

32

D A V I D L 0 P E Z and C Y N T H I A

FE LI C I A N 0

those producing these babies. The majority of the Latinos entering California's labor force after about the year 2010 will have grown up in California, the products of California's benighted public schools. It is one thing to dismiss the state's responsibility to immigrants by saying that they could have stayed home. But for the Latino labor force that will emerge in the next century, California will be the only home they know.

Ethnicity and Nativity at Work We turn now to California's workforce, defined here as those aged 25-64 who are in the labor force, amounting to about thirteen million workers in 1996. 7 Figure 1.4 shows the absolute size of the ethnic components of California's work force from 1970 to 2000, and is quite important in several respects. It reveals that all four major ethnic components grew, though at different rates. The state's entire workforce more than doubled, from 5,955,000 in 1970 to an estimated 13,100,000 in 1996. Over the same period the white labor force kept pace up to 1990, after which time it has been essentially static. Even the African American workforce grew faster than the white, but rates of increase for Latinos and Asians were much greater. From 1970 to 1996 the black workforce grew 158 percent, but the Latino workforce grew about 500 percent and the Asian about 800 percent. The evidence of figure 1.4, combined with the ethnic components of growth shown earlier, suggest that the state's workforce will eventually follow the population as a whole, with Latinos emerging as the single largest segment. But this will not happen until well into the next century. Table 1.1 displays the ethnic I nativity composition of the principal occupational groups in California's workforce in 1996. 8 Whites still dominate the workforce overall, but the next two largest components are foreign-born Latinos and foreign-born Asians, who together constitute 27 7. Excluded are the 3.5 million in this age band who are not working, as well as younger and older workers. Probably the most serious exclusion is of workers under age 25, who of course are present in abundance in California's shops and factories. The argument in favor of beginning with 25-year-olds is that by that age most people have completed their education and are in jobs similar in status to what they will do the rest of their lives, thus allowing for better comparisons across ethnic and generational lines. But of course poorer folk, most notably Latino immigrants in this case, start working well before age 25, and the Latino population is also on average younger. Just as census figures disproportionately undercount poor Latinos, we also know that excluding young workers disproportionately undercounts poor Latinos. Our estimates of the Latino presence in various jobs and industries, then, are conservative. 8. We have not distinguished between native and foreign-born whites or African Americans, since immigrants constitute only 8 and 6 percent respectively.

Who Does What?

33

8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 1970 DWhite

1980 0 Latino

1990 •Asian

2000 • Black

Sources: 1970-1990 U.S. Census of Population, Integrated Public Use Sample (IPUMS-95), Version 1.0. Year 2000 estimates based on 1997 Current Population Survey and estimates from California Department of Finance, California Demographics (Winter 1998).

Figure 1.4. Size and ethnicity of California workforce, 1970-2000 (in thousands). percent of the workforce (if white and black immigrants were added in, the total would be 33 percent foreign-born). Over two-thirds of the Latino workforce is foreign-hom, as is over 80 percent of the Asian workforce. Indeed, immigration from Asia and Latin America in the past thirty years has overwhelmed the older Latino and Asian second and third generations that, as recently as 1970, constituted half the Asian adults and 60 percent of Latino adults in California. The occupational distribution of Latino and, to a lesser extent, Asian immigrants contrasts sharply with that of the (largely native) white group, as table 1.1 shows. Whites constituted 56 percent of all workers in 1996, but their representation in different occupational levels varied considerably. They made up 71 percent of all professional, managerial, and technical workers, but only 30 percent of all laborers. In marked contrast, immigrant Latinos constituted 17 percent of all employed Californians but 42 percent of the factory workers (operatives), 49 percent of the laborers, and only 5 percent of the professional and technical employees. The occupational distributions are less polarized for all the other groups:

DAVID l 0 P E Z and CYNTH I A FEliCIANO

34

Table 1.1. Ethnic and nativity composition of workforce aged 25-64, California, 1996 (percentage) . White Black Professional I technical workers Clerical I sales workers Crafts workers Operatives

Latino native

Latino foreign

Asian native

Asian foreign

Total number (%of total)

71%

5%

6%

5%

3%

10%

4,763,700 (36%)

59%

8%

10%

9%

3%

10%

54%

5%

10%

20%

2%

8%

32%

4%

10%

42%

1%

11%

Laborers

30%

5%

9%

49%

1%

5%

Service workers Total

38%

7%

8%

36%

1%

9%

56%

6%

8%

17%

2%

10%

3,285,600 (25%) 1,574,200 (12%) 1,242,800 (9%) 869,300 (7%) 1,385,400 (11%) 13,120,000 (100%)

Source: 1995-1997 Current Population Survey.

Mrican Americans constitute 4-8 percent of each major occupation, native Latinos 6-10 percent of each, immigrant Asians 8-11 percent of every occupation except for laborer (5 percent), and native-born Asians 2-3 percent of white-collar and skilled workers, though only one percent of the semi- and unskilled workforce. 9 The two most ethnically distinct major occupational categories, then, are at the top (whites 71 percent of the professional and technical workers) and bottom (Latino immigrants 49 percent of laborers). Within these broad categories, of course, are specific occupations (Hollywood producer, grape picker) that are even more ethnic-specific. But instead of pursuing such occupational specifics, in table 1.2 we shift the focus to the ethnic groups themselves and their broad occupational and income profiles.10 Looking first at whites, blacks, and native-born Latinos, we find the 9. Space limitations prohibit us from presenting gender breakdowns. For the most part the usual gender disparities cut across lines of ethnicity and nativity. That is, in most specific groups women are less likely to be skilled workers or laborers and correspondingly more likely to be clerical or service workers. The most striking gender disparities come not in occupational level but in income. 10. Table 1.2 is derived from 1990 Census data rather than from the 1995-97 Current Population Survey data on which Table 1.1 is based, in order to increase sample sizes for the smaller ethnicity and occupation categories.

Who Does What?

35

familiar white I nonwhite disparity: nearly half of all white workers are found in managerial, professional, and technical positions, compared to less than a third of Mrican Americans or native Latinos. Conversely, the latter two are about twice as likely to be factory or service workers. White/nonwhite income gaps are also substantial: whites earn about 25 percent more overall and maintain a substantial advantage at each occupational level except for service, where black workers earn slightly more. This anomaly is probably due to the tendency of blacks to be in the public sector, where service work is paid substantially better than in the private sector. The occupational profile for U.S.-born Asian Americans is similar to that for whites, and is in fact somewhat elevated: Asian Americans have the highest proportion of professional, managerial, and technical workers and the lowest proportion of factory and service workers. In terms of income, they are similar to whites by occupational level and overall. Turning to the two immigrant groups, we see two very distinct profiles. Latino immigrants provide the greatest contrast to whites and native Asian Americans: only 13 percent are in the top occupational category, and they are three to five times more likely than whites or native Asians to Table 1.2. Occupation and earnings of workforce aged 25-64 by ethnicity and nativity, California, 1990. White

Professional I managerial I technical 47% (Mean earnings, in thousands of dollars) ($36.1) Clerical I sales 25% Craft Operatives Laborers Service Total

Black

Latino native

Latino foreign

Asian native

Asian foreign

32%

28%

13%

52%

40%

($27.6) 27% ($20.1) 14% ($27.1) 13% ($21.3) 6% ($18.4) 11% ($12.4) 100% ($22.4)

($22.9) 12% ($15.9) 14% ($18.2) 25% ($13.7) 16% ($11.6) 20% ($97.6) 100% ($14.6)

($35.1) 26% ($23.6) 8% ($30.4) 5% ($21.3) 3% ($19.2) 5% ($16.0) 100% ($29.4)

($31.3) 24% ($18.4) 8% ($22.1) 12% ($15.5) 2% ($13.9) 13% ($12.5) 100% ($22.6)

($29.0) 30% ($23.6) ($21.1) 10% 12% ($28.6) . ($26.9) 10% 6% ($23.4) ($22.1) 3% 4% ($20.5) ($17.9) 7% 15% ($13.9) ($14.3) 100% 100% ($29.2) ($23.2)

Source: 1990 U.S. Census of Population, Public Use S=ple, Version 1.0.

36

DAVID LOPEZ and CYNTHIA FELICIANO

be factory workers, laborers, or service workers. Overall, 48 percent of whites and native Asians are professional, managerial, or technical workers, and only 15 percent are laborers, factory workers, or service workers; the occupational profile for immigrant Latinos is just the reverse with 61 percent found at the three lowest occupational levels, compared to only 13 percent at the top. Furthermore, immigrant Latinos have the greatest income disparities at each occupational level and, of course, overall. The overall disparity (they earn about half what whites and native Asians do) is hardly surprising, but the magnitude of the disparity at each occupational level is worth calling attention to; at every level Latinos earn only about two-thirds what whites do, and also substantially less than any other group. The reasons for these disparities are not difficult to discern: most Latino immigrants arrive with low levels of education, poor English skills, and little or no financial capital. They average only eight years of schooling and only 14 percent have any education above high school (Current Population Survey 1996). Averages do obscure individual differences: among immigrant Latinos are to be found millionaire surgeons and, more to the point, a substantial minority that have earned middle-class or "affluent working-class" status through dint of hard work and thrift. But the fact remains that two-thirds of all immigrant Latino workers are in low-paying, low-status "dirty-work" jobs. Asian immigrants provide a more mixed picture, in some ways resembling whites and native Asian Americans, and in other ways resembling Latino immigrants. Forty percent are in professional, managerial, or technical occupations, more than African Americans or native Latinos. Like the latter, they are more likely than whites to be found in factory or service work and are markedly underrepresented among laborers. This mixed occupational picture reflects the fact that recent Asian immigrants are probably the most diverse group in the history of immigration to the United States. About two-thirds come from middle-class backgrounds and arrive with high levels of education and good English skills. The other third have more humble backgrounds, such as the second wave of Southeast Asian refugees who were largely peasants, fisherfolk and small-town workers. Overall, Asian immigrants average thirteen years of schooling, close to the U.S. national average, and 62 percent have had some college education. In ways unique in U.S. immigration history, the well-equipped two-thirds move directly into professional, technical, and commercial occupations at or above the U.S. average.U 11. Schoeni, McCarthy, and Vemez (1996) examine the segments of the immigrant workforce in more detail, and attempt to assess their future prospects.

Who Does What?

37

But the Asian immigrant experience today is by no means a uniform picture of smooth and rapid economic assimilation. High averages mask the struggles of the poorly educated minority, and some Asian immigrants, especially those with relatively modest English skills, settle for careers well below average for their level of education, Korean shopkeepers being the best-known example (Cheng and Yang 1996). And, as table 1.2 shows, at every occupational level their earnings are below whites' and native Asians', though always above immigrant Latinos'. This is true even at the lower rungs of the occupational ladder, suggesting that even the less well-prepared Asian immigrants are doing substantially better than their Latino counterparts. Yet, though blacks and native-born Latinos are somewhat more highly represented in manual occupations than Asian immigrants, the wages of the latter are substantially lower, suggesting that they, like immigrant Latinos, are found in lower-paying jobs and industries. Table 1.3 provides an overview of the industries in which each group is found. By far the greatest immigrant concentration is in agriculture, where fully 62 percent of all those employed (not just laborers) are foreign-born Latinos, nearly four times their proportion of the total state workforce. They are also slightly overrepresented in construction, though that comparatively well-paying industry remains predominantly white. Whites also continue to predominate in the best-paying and most whitecollar of industries, FIRE (finance, insurance, and real estate), where immigrant Latinos are the only group that is substantially underrepresented. In manufacturing, by contrast, the two most overrepresented groups are immigrant Latinos and Asians. In personal service, immigrant

Table 1.3. Major California industries by ethnicity and nativity, 1996.

White Black Agriculture Construction FIRE Manufacturing Personal service Trade Transport Other Total

28% 62% 63% 45% 48% 54% 52% 56% 56%

1% 4% 7% 5% 5% 4% 10% 8% 6%

Latino native

Latino foreign

6% 9% 8% 8% 7% 9% 12% 8% 8%

62% 20% 7% 25% 27% 20% 12% 15% 17%

Source: 1995-1997 Current Population Survey.

Asian

Asian

native

foreign

Total number

2% 1% 1% 2% 4% 2% 2%

3% 4% 12% 15% 11% 11% 10% 10% 10%

435,000 817,900 835,100 2,280,300 435,000 2,230,600 569,600 5,064,300 13,168,000

38

D A V I D L 0 P E Z and C Y N T H I A F E L I C I A N 0

Latinos are the only group that is overrepresented. 12 Both trade and transport have ethnic profiles approaching the overall average. Comparing tables 1.2 and 1.3, we see that ethnic disparities are more substantial by occupational level than by industry. Agriculture is the only major industry that has become predominantly Latino. Not coincidentally, agriculture is also the only major industry where laborers are the single largest occupational category. We noted earlier that Los Angeles and San Francisco have been affected by immigration in very different ways. Table 1.4 profiles the Latino and Asian immigrant and native white workforces in these two metropolitan areas, with comparisons to the state as a whole. Perhaps the single most important contrast is in relative size: Latino immigrants are 19 percent of the Los Angeles workforce, but only 6 percent in the Bay Area. Asian representation in the two workforces is much closer: 8 and 11 percent respectively. But the more important comparison is within each region: Latino immigrants are the single largest category of nonwhite workers in Los Angeles, and they are radically behind whites and Asians in educational, occupational, and income attainment. In San Francisco, Asian immigrants are the single largest category of nonwhite workers and, though they are not at parity with whites, in most important respects they approach them. (The one exception is the minority of immigrant Asians with little or no schooling. Though only one-tenth of the total, those with this educational level are virtually nonexistent among any ethnic group that has grown up in the United States.) The most consequential differences between Los Angeles and San Francisco show up in the occupational distributions of each ethnic group. In Los Angeles, factory work ("operatives" in the table), service, and laborer are all identifiably Latino occupations, whereas in San Francisco the occupation with the greatest Latino representation is laborer, at 26 percent, and even here Latinos are substantially outnumbered by whites, who constitute 43 percent. These basic demographic differences go a long way toward explaining the less conflictual nature of racial/ ethnic relations and politics in the Bay Area. In San Francisco the largest and most visible nonwhite immigrant group is at rough economic parity with whites; in Los Angeles the gap between whites and immigrant Latinos is a significant disparity. The next three tables map out the industrial distribution and ethnic composition of the three broad manual occupational levels in 1990: 12. Law enforcement and some other comparatively well-paying service industry subcategories are excluded from personal service.

Who Does What?

39

Table 1.4. Immigrant Latino and Asian and native white workforces aged 25..:.64 in Los Angeles, San Francisco Bay area, and state of California, 1990. Latino immigrants Calif.

L.A.

S.F.

Calif.

L.A.

S.F.

Calif.

19% 6% 48% 37% 17% 29% 23% 29%

17% 48% 14% 25%

8% 10% 65% 69%

11% 11% 63% 62%

10% 10% 62% 64%

48% 1% 68% 74%

61% 1% 74% 75%

58% 1% 67% 72%

59% 58% 42% 13% 48% 26%

61% 42% 49%

22% 29% 7% 17% 4% 7%

27% 11% 5%

15% 28% 27%

15% 44% 43%

16% 36% 34%

33% 17%

36%

8% 16%

9%

35%

43%

40%

72% 37%

54%

5% 12%

6%

11%

30%

27%

Average male earnings 17.7 21.5 (thousands ofU. S. dollars) Source: 1990 U.S.

White natives

S.F.

L.A.

% oflabor force % < 9 years school % any college % white collar % operative I labor I service % of operatives % oflaborers %ofservice workers % of household workers

Asian immigrants

32.0 30.8

44.9

44.5

Census of Population, Public Use Sample, Version 1.0.

operatives, laborers, and service workers. Each table includes the pro-portion of the occupation that is immigrant Latino or Asian, the proportion reporting that they entered the United States after 1985, and an "equity index" for each group of immigrant workers, which is their average earnings divided by the average earnings· of all workers in the same occupation. Table 1.5 profiles California's immigrant operatives. The core of this broad occupational category is semiskilled manufacturing workers, though half of all operatives are outside this sector, spread across a myriad of trade, transport, and other industries. Within manufacturing, foreign-born Latinos constitute 41 percent of all operatives and immigrant Asians 14 percent. Thus these two nonwhite immigrant groups constitute fully 55 percent of all the manufacturing operatives in the state in 1990, a proportion that surely has continued to rise. In apparel manufacturing, nonwhite immigrants were 89 percent of all operatives, and in textiles, 78 percent. But if the apparel and textile industries in California are totally dependent on immigrant labor, the reverse is not true: these industries employ only about 14 percent of all immigrant ope,ratives.

40

D A V I D L 0 P E Z and C Y N T H I A

FELI C I A N 0

Table 1.5. California's operatives by ethnicity, nativity, and industry, 1990.

Foreign-born Latinos Foreign-born Asians Number 1,119,636 Total Manufacturing 558,737 Apparel 61,519 Textiles 16,840 Machinery 124,237 Food 53,181

Equity Equity Percentage index' Percentage index 32% 41% 64% 67% 30% 39%

.72 .74 .95 .97 .83 .72

10% 14% 25% 11% 27% 6%

.81 .86 .98 1.03 .89 .78

% newcomerb (Latinos only) 19% 25% 20% 16% 14%

Source: 1990 U.S. Census of Population, Public Use Sample, Version 1.0.

' The equity index is the average earnings of the ethnic group divided by the overall earnings in the specified occupation and industry. b Newcomers are foreign-born persons who report they entered the United States after 1985.

The equity indices show that Latino and, to a lesser extent, Asian immigrant wages lag behind others, even at this level of occupation I industry specification. In apparel and textiles, where the vast majority of the workers are Latino and Asian immigrants, the index means very little, but the substantial disparities in the machinery and food processing industries are more significant, all the more so because these fields have comparatively low proportions of newcomers. The highest proportion of newcomers, 25 percent, are found in apparel. 13 Table 1.6 provides similar information for the state's laborers. Though a substantial portion of Asian immigrants do become factory workers, they are less likely than average and much less likely than Latino immigrants to work as laborers. This is especially true in agriculture, where labor is poorly paid, but it is also true in construction, where at least some laborers are well remunerated. As expected, the greatest immigrant Latino concentration is in agriculture (63 percent of all laborers). In construction, however, only about one-third of all laborers are Latino immigrants, reflecting the continuing dominance of this relatively well-unionized industry by whites. Ethnic inequalities in income are particularly great for construction laborers, probably because of unequal rates of unionization between whites and Latinos. Overall, immigrant 13. The true figure is almost certainly higher, since this is an industry 11~th a large proportion of undocumented workers who either evade the census and survey takers, or who have reason to report they have been in the United States longer than they actually have been.

Who Does What?

41

Table 1.6. California's laborers by ethnicity, nativity, and industry, 1990. Foreign-born Latinos Foreign-born Asians Number Total 531,418 Agriculture 189,578 Construction 114,838 Trade 86,692 Manufacturing 51,137

Equity Equity Percentage index• Percentage index 41% 63% 35% 23% 38%

.73 .89 .71 .76 .76

4% 3% 2% 6% 6%

.87 .81 .83 .87 .92

% newcomerb (Latinos only) 21% 24% 20% 19%

Source: 1990 U.S. Census of Population, Public Use Sample, Version 1.0. ' The equity index is the average earnings of the ethnic group divided by the overall average earnings in the specified occupation and industry. h Newcomers are foreign-born persons who report they entered the United States after 1985.

Table 1. 7. California's service workers by ethnicity, nativity, and industry, 1990. Foreign-born Latinos Foreign-born Asians Number Total Household Hotel Laundry Trade

1,131,584 79,630 61,335 82,490 292,897

Equity Equity Percentage index• Percentage index 25% 54% 42% 9% 30%

.78 .91 .85 .90 .89

10% 6% 15% 11% 15%

1.0 1.3 1.1 0.9 0.9

% newcomerb (Latinos only) 9% 18% 13% 3% 13%

Source: 1990 U.S. Census of Population, Public Use Sample, Version 1.0. • The equity index is the average earnings of the ethnic group divided by the overall average earnings in the specified occupation and industry. b Newcomers are foreign-born persons who report they entered the United States after 1985.

Latino laborers earn 73 percent of the average wage, about the same as among operatives. Of the four broad industries shown, Latino laborers in construction are somewhat more likely to be newcomers, but the rate of recent arrivals is 19 percent or more in each industry. Table 1. 7 describes the composition of service workers. These workers are spread throughout California's economy, with the greatest concentrations in trade and personal services (broken down into its three principal components: household services, hotel work, and laundry work). Overall, Asian immigrants are in service occupations roughly in proportion to their total representation in the labor force (10 percent); however, they

42

D A V I D l 0 P E Z and C Y N T H I A

F E ll C I A N 0

are underrepresented in private household work. In contrast, Latino immigrants constitute 25 percent of all service workers (they make up 17 percent of the total state workforce), holding over half of all household service jobs and 42 percent of all hotel service jobs. Of the three most stereotypical immigrant jobs-agricultural laborer, garment factory worker, household worker-the latter is the most difficult to get at using the broad occupation I industry categories employed in this chapter. Part of the reason is that household service, though growing again, is still a comparatively small occupation at the statewide level. It may also be the most underreported occupation, since household workers are, by all accounts, the most undocumented workers in the state and also probably the best hidden from census takers. Studies of Central Americans in southern California document that Salvadorans and Guatemalans are especially likely to be household workers, and of course the vast majority are women (Lopez, Popkin, and Telles 1996, 296). Income inequality is relatively modest within service occupations. But it is worth pointing out that these are some of the lowest-paying jobs in the state. The average wage for private household workers is only $7,297; for Latinas it is $6,609. Asian immigrant service workers are paid at least average or somewhat above average in these low-wage jobs. Immigrant workers in service are among the least likely to be recent arrivals to the United States, with overall rates about half that for laborers and factory workers. The service job that is most associated with immigrants, private household work, has a newcomer rate approaching that for labor or factory work, and considerably higher than that for other service jobs. But household workers constitute only 7 percent of all service jobs. For other service work, it may be that the combination of language skills and contacts needed to find jobs ensure that they remain the province of better-settled immigrant workers.

A Glimpse at the Future: The Emerging Second Generation

The previous section was devoted to a cross-sectional overview of the situation of immigrant workers in California. But of course the overwhelming fact about California's workforce is change, not stasis. The future course of immigration to California is impossible to predict, but it will probably continue pretty much as before: poorly educated and nonEnglish-speaking immigrants from Mexico and Central America will continue to be available for the least desirable jobs throughout the economy, and diverse but generally much better-educated Asian immigrants will enter the job market up and down the occupational pyramid. Put an-

Who Does What?

43

other way, Latin American immigration will continue to contribute to the appearance of a castelike social system. In contrast, Asian immigration, to the extent that it continues to be socioeconomically diverse, actually has the potential to reduce the association between class and ethnicity. Whatever the future may hold with regard to immigration, in the next thirty years the children of immigrants will enter the labor force in numbers equal to or above the number of immigrants in the past thirty years. Where will they fit in? The answer is only beginning to emerge, and is fraught with complexities beyond the scope of this discussion; but we can offer a preliminary guess. 14 California's high school students of today will constitute half or more of its young workers a decade from now, the only other significant source being immigrants from Latin America and Asia (a reasonable assumption given the low rates of in-migration from other states in recent decades). Studies have shown that the relative performance of young people in school provides a good guide to where they will end up in the occupational I class structure in the future, both across ethnic groups and within them (Blau and Duncan 1967). Information about ethnic stratification among the state's schoolchildren, then, provides important hints about the future ethnic stratification of its workforce. School data are rarely reported by generation, but there is a remarkable correspondence between ethnicity and generation among highschool-age youth in California today: over 90 percent of Asians are second or "1.5" generation (born abroad but schooled in the United States), as are over 80 percent of Latinos, whereas the corresponding figure for whites is under 20 percent and for blacks under 10 percent. To a considerable degree, then, high school students divided by ethnicity are also divided by generation. How are California's ethnic groups doing in school? The more usual question today is, how are the schools doing? From most perspectives the answer is, poorly. But hidden behind the subpar average performances are significant ethnic differences, which are only partially captured by ethnic variations in high school graduation rates and average years of schooling. For decades Latinos have had low high school graduation rates, and these rates have not varied much by generation. The most recent data available, from the 1996 Current Population Survey, tell much the same story. By one common measure, the percent aged 16-21 who are either high school graduates or still in school, whites, Asians, and Mrican Americans are bunched together at about 90 percent, with no sig14. See Portes and Zhou ( 1993) for a comprehensive discussion of the factors affecting the assimilation of immigrant children in the United States today.

44

D A V I D L 0 P E Z and C Y N T H I A

F ELI C I A N 0

nificant variation by generation. Latinos are 10-12 points below this level, again with no significant generational differences. 15 As high school graduation is virtually universal (and perhaps more debased in value than in the past) and because most good jobs demand more in the way of qualifications, it is a poor indicator of group differences. Years of schooling is only a little better as an indicator, for most groups bunch up around 12-14 years on average, and in any case this measure can be applied only to age groups who have completed their education. In California there is, however, an excellent measure of the likelihood of entering the middle and upper-middle classes, for groups if not for individuals: eligibility for the University of California (UC) .16 In theory the top 12.5 percent of high school graduates are eligible for UC, though in fact about 20 percent are, by a combination of grades and test scores. 17 Less than half actually attend a UC campus, but it is a safe bet that most of this "talented two-tenths" do go on to college; they probably constitute the majority of those students who attend and graduate expeditiously from the state's second public university system as well the majority of private college enrollment. · Table 1.8 gives the high school graduation rates and UC-eligible rates by ethnicity, as well as the ethnic composition of the high school graduating class and the pool of those potentially eligible for the University of California. The total eligible figures are the key here: they suggest that whites will be three times more likely to be in the state's future elite than Latinos or Mrican Americans, and Asians will be six times more likely. Put another way, 44 percent of Asian graduates are headed for the state's elite, compared to 8 percent of Latino graduates. These figures are re15. The very low rate~ of high school graduation sometimes noted for Latinos stems from lumping immigrants with natives; when only those raised in the United States are compared, the Latino rate is substantially higher, though still below average. 16. Of course there are other standards. Test scores are an obvious one, but until recently comprehensive and comparable scores for students across the state were not available. With the recent institution of the widely used "Stanford Nine" as the state's test of achievement for grades 2-11, that is changing. Explicit ethnic breakdowns are not yet available, but preliminary estimates made by the senior author suggest that ethnic differentials will be quite considerable,. with Latino and black students averaging in the twentieth to thirty-fifth percentile, whites between the fortieth and fifty-frfth, and Asians between the frftieth and seventieth percentile on the various grade and subjectmatter tests. 17. Currently only about 11 percent meet all the technical criteria for entry, which include taking an SAT subject exam, something not required by many other colleges and universities. Ethnic disparities are similar among students technically eligible, and among those who actually attend the University of California. We use the broader category of all who are potentially eligible because we believe it better represents the state's academic elite.

Who Does What?

45

Table 1.8. Educational achievement among California youth by ethnicity, 1995. All high school graduates (as% of all aged 16-21)

White Asian Latino Black

92% 89% 82% 90%

%of all high school graduates 43% 13% 38% 6%

-100%

%potential UC-eligible

%of all UCeligible

26% 44% 8% 7%

50% 28% 14% 2% 100%

Sources: High school graduation rate (students aged 16-21 graduated and/ or still in school) from 1995-1997 Current Population Survey; UC eligibility data from California Post-secondary Education Commission (CPEC).

markably similar to the proportion of each group of immigrants who have professional, managerial, and technical jobs (see table 1.2). They suggest that inequality in the second generation will be massive, equivalent to the disparities between the two major sets of adult immigrants. The final set of figures represents the composition of the state's academic elite (top 20 percent): half are white, 28 percent Asian, 14 percent Latino, and only 2 percent black. These figures are a rough guide to the ethnic composition of the top quarter of the state's future occupational hierarchy. Many factors could intervene to moderate these harsh contrasts: the correlation between high school grades and eventual status is not perfect. Furthermore, the Latino second generation will certainly experience social mobility in contrast to their parents, from the bottom to the middle rungs. But to the degree that upward mobility into the top half of the occupational structure requires successful competition with whites and Asians, it is not in the cards for most of the Latino second generation. Their future will depend not so much on individual achievement but rather on the quality of life for those in the middle of the state's class structure. We can only make educated guesses about the future of the young people represented in table 1.8. Table 1.9 looks at a cohort about ten years older, those aged 25-34 who are in the workforce and largely finished with their formal education. It suggests that ethnic disparities in this generation may not be so great as those suggested by comparing immigrant Latinos with others, or by comparing today's Latino (and black)

46

D A V I D l 0 P E Z and C Y N T H I A F E ll C I A N 0

Table 1.9. Educational and occupational status of native-born California workers aged 25-34 by ethnicity, 1996.

%some college

%college graduates

%whitecollar

%craft

% WISkilled

67% 73% 53% 57%

32% 48% 16% 16%

67% 75% 57% 62%

14% 12% 16% 13%

19% 13% 27% 25%

White Asian Latino Black

Source: 1995-1997 Current Population Survey.

high school graduates with other ethnic groups. By the demanding standard of college graduation, the ethnic disparities are considerable: nearly half of all young native Asian workers are college graduates, as are onethird ofwhites. But only 16 percent of native Latino (or black) young workers have a college education. The disparities diminish when the standard is having any post-secondary education. Latinos are 14 percentage points behind whites and 20 points behind Asians in the proportion with some college education. These data correspond closely with the percentage who are in white-collar occupations. The converse is that Latinos are substantially more likely to have unskilled jobs, about twice as likely as Asians. In short, these educational data suggest that social disparities among ethnic groups growing up in California today will be sharpest at the upper levels, and possibly greater than the differences observable today between whites and Asians on the one hand and Latinos and African Americans on the other. At the lower end of the spectrum, perhaps onequarter of U.S.-raised Latinos (and African Americans) will end up as unskilled workers, 50-75 percent more than will be the case for Asians and whites. U.S.-raised Latinos will form an increasing proportion of those with jobs in the middle, the lower-status white-collar jobs and crafts. Or, to summarize in less optimistic terms, competition with better qualified whites and Asians will continue to keep them out of the top tier, and competition with more desperate Latino immigrants will keep them off the bottom; one can only hope that the jobs will be there for them in between. Conclusion

Immigrants arrive in the United States with a broad array of backgrounds and talents, and they find or create jobs that range across the entire oc-

Who Does What?

47

cupational spectrum. But they do not simply blend into the native workforce: about half do, but the other half are without the educational, linguistic, social, and financial resources needed to move into jobs with a future. Not all of the latter are from Mexico and Central America, but the vast majority are, just as the m,Yority of the more fortunate immigrants are from Asia. The focus of this chapter has been on immigrant workers who fill the semi- and unskilled jobs in California's booming economy, and our goal has been the simple one of describing where immigrant workers are most likely to be found. We have no~ dealt directly with the contentious question of their contribution to this boom, but it is virtually certain that these low-wage workers have contributed to the well-being of other Californians, caring for their children and their lawns, cooking their food, sewing their jogging suits, and operating the machines that stamp out a myriad of inexpensive consumer goods. Nor have we considered the question of intragenerational mobility among immigrant workers, though the low wages and occupational status that persistently characterize Latino immigrants suggest that social mobility is modest for these immigrants during their lifetimes. We have taken a first pass at a third question, which we consider to be especially significant: the educational and occupational prospects of the vast "new second generation," the children of today's immigrant workers. About one-third, including most Asians and a minority of Latinos, seem likely to end up at least as well off as native-born whites. But regarding the other two-thirds, largely low-achieving students from struggling and overwhelmingly Latino immigrant homes, there is cause for concern. The previous Latino second generation hardly excelled at school, but entered a workforce that provided good, often unionized, jobs to many. But changes in the economy, plus the vastly greater size of the new second generation, make it difficult to predict that they will end up even as well off as today's Chicano working class. If even a significant minority of the new second generation do in fact join today's small urban "underclass," the social consequences may well be cataclysmic. Our conclusions on this point must be tentative, as most of the new second generation are still too young to be in the workforce, but evidence from school years and early entry into the job market provide little basis for optimism. Whatever the fate of the second generation, we can be certain that immigrants will continue to come, eager to take up the jobs others disdain at wages that are miserable by U.S. standards but also five times the prevailing wage in Mexico. Industries such as apparel, food, other services, most low-tech manufacturing, and, most important of all, agriculture, will

48

D A V I D L 0 P E Z and C Y N T H I A F E L I C I A N 0

continue to be dependent on immigrant workers, and specifically on the continued arrival of fresh new recruits from across the border. Middleclass immigrants will integrate more or less smoothly into the top half of the information economy, joining the largely white and Asian middle class. The low-achieving second generation will be sandwiched somewhere in between. As for low-skilled immigrants, eager for any wages at all, they will continue to do the dirty work in all sectors of the economy, from the fields to the high-tech factories of Silicon Valley. We can only hope that at least some of them will be working under the protection of unions.

2

Immigrant Workers and American Labor: Challenge ... or Disaster? Roger Waldinger and Claudia Der·Martirosian

S

omnolent for much of the past twenty-five years, American labor approaches century's end with new vigor. And not a moment too early, as the problems are legion. Unionization rates are down and still falling. Downward pressure on wages has not yet abated-even though the U.S. economy is enjoying a boom of unusual duration. The labor force is growing at galloping rates-making it hard for labor to simply tread water, let alone move ahead. And much of the job proliferation takes place in those sectors where unions have historically been weak, a situation aggravated by the fact that many of the new labor force entrants have uncertain, often transient connections to the organizations that employ them. True, unions have learned a few quite potent new tricks, which they have used to some effect. But neither the arsenal of labor's weapons nor the political environment is such as to force employers to abandon the bad habits they acquired during the years of labor's decline. So the vicious cycle continues, with the steadily diminishing union presence at once weakening labor's appeal to potential members and increasing employers' motivation to resist organizing efforts. _ Daunting as the situation may seem, the demographic transformation of America confronts unions with yet another challenge-the renewal of mass immigration. The 1930s and 1940s are known as the heyday of American labor; viewed from another perspective, they also encompass a period of highly restricted, indeed virtually negligible immigration. As such, labor's years of triumph represent a twentieth-century anomaly. ImThanks to the Cniversity of California's Mexus Program for support for the research on which this chapter is based.

49

50

R 0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R 0 S I A N

migration largely flowed unimpeded during the century's first two and a half decades, the World War I years excepted. And though the barriers put in place in 1924 were never entirely removed, the post-World War II turnaround is unmistakable, constant, and, in recent years, immense. If not record-breaking, immigration levels during the 1990s will fall short only of the total recorded during the first decade of this century; the 1980-2000 totals will significantly exceed the influx experienced during 1900-1920. Of course, correlation does not imply causality, as any social scientist knows all too well. Still, the historical record does suggest an uncomfortable relationship between America's high-water years of immigration and a flagging labor movement. And one only need look at the experience of the United Farm Workers, their ranks decimated just when immigrant labor surged through California's fields, to glimpse what immigration may have in store for organized labor. If the picture is not hopelessly bleak, it certainly presents a somber aspect. The new "new immigration" of the late twentieth century is not as uniformly proletarian as the "new immigration" of old, but the numbers of new entrants moving in at the very bottom is impressive indeed (see Lopez and Feliciano, this volume). Convergence at the low end of the labor market both increases downward pressure on wages for less-skilled labor and adds to the looseness of the labor supply-neither of which developments gives labor any additional leverage. Many of the new immigrants are recent arrivals, altogether too uncertain of their status, standing, and orientation-will they settle down or return home?-to seriously consider unionization. Moreover, a substantial portion of the immigrant population-as many as one out of every four immigrants arriving each year-has an additional fear factor, associated with a condition rarely known earlier in the century, namely, a status prohibiting work in the United States. These "illegal" or "undocumented" immigrants-call them what you will-know all too well that the Immigration and Naturalization Servic.e provides employers with an additional union-busting tool, yet another reason to desist from organizing activities.' Furthermore, the immigrants tend not to move into those industries where the labor presence remains entrenched-most notably, the public sector, where a varying combination of citizenship, educational, and testing requirements exclude many foreign-born workers. Instead, the newcomers head for notoriously union-resistant industries such as apparel or restaurants, where the addition of a seemingly limitless supply of labor makes organizing all the more difficult. 1. However, undocumented status may not prove an insuperable obstacle to unionization; see Delgado 1993.

Immigrant Workers and American Labor

51

But immigration does not necessarily spell bad news. In the past, many union leaders emerged from immigrant ranks; the history of American labor also includes countless extraordinary eruptions of immigrant militancy. And it is not simply hopeless nostalgia that suggests that the past may be prologue. Assimilation is a time-honored concept in the study of American immigration and now would not be the first time that "Americanization" involves a process whereby immigrants decide that they want-and deserve-the good things promised by American life. 2 If the newcomers have also acquired "a diploma in exploitation," to quote a rank-and-file immigrant janitor union leader (see Waldinger et al. 1998), and have come to resent the maltreatment and stigmatization experienced at the hands of employers and society alike, then immigrant workers may decide that unions provide them with a powerful instrument of collective voice. From a historical point of view, one might also contend that it is labor, not the immigrants, that deserves the bad reputation. Few organizations could have rivaled the old AFL in its impassioned restrictionism-good reason for the immigrants of yore to be skeptical, if not skittish, about appeals to solidarity. This time around, however, organized labor does not seem likely to make the same mistake. Indeed, the AFL-CIO is one of the few mainstream organizations to have emerged as decidedly proimmigrant, a stance marked not just by rhetoric but by action. As evidence, consider the efforts to organize America's newest workers, many of which are catalogued and analyzed in this volume. This chapter offers an overview of the relationship between immigrant status and unionization, both nationally and in California, the capital of today's immigrant America. We analyze the factors affecting the likelihood that a worker will have a "union job," which we define as someone who is a union member or working in a unionized establishment. The occasion is the availability of a new data source, resulting from therecent adoption of questions about place of birth and parentage by the monthly Current Population Survey (CPS), an instrument which also collects information about union membership and employment in establishments covered by union contracts. Though the CPS does offer a rich source of information, we should issue some cautions on its limitations before proceeding any further. Most important, the data shed light on the factors associated with union membership among immigrants and native-born workers, a matter quite distinct from the factors that would influence immigrants and natives to join unions at differential rates. The 2. As in the earlier CIO period studied by Cohen (1990).

52

R0 G E R W A L D I N G ER and C LA U D I A D ER- M A RT I R0 S I AN

CPS provides a repository of information on demographic characteristics; however, it contains neither attitudinal data that would measure preferences for union membership nor relevant behavioral data conceming voting behavior in elections for union representation. One can argue, as some researchers have, that membership status can be equated with the "demand for union services," and thus serves as a reasonable proxy for the attitudinal and behavior indicators of interest (Duncan and Stafford 1980). Survey research does provide some support for this point of view, as union members appear to be significantly more likely than their nonunion counterparts to vote union in a hypothetical election for union representation (Leigh 1986). In the current context of declining union membership, moreover, where there are more persons wanting union jobs than union jobs available, those with such jobs would seem highly likely to prefer whatever benefits unionization generates. Though this asymmetry between the supply and demand for union jobs also implies a considerable interest in unionization among the nonunionized workforce (as survey evidence indicates), here we examine patterns of union membership among persons holding "un1onjobs" by ethnicity and nativity. Second, the CPS inquires broadly about labor unions and employee associations, not differentiating between the two; the membership information that it obtains is of a highly general sort and cannot be used to generate estimates of the membership of particular unions. Third, the CPS takes a cross-section of the labor force, a complication for those arguments about unionization that involve change over time. As we've already suggested, a variant of the assimilation hypothesis would suggest that propensities to unionize increase over time, a hypothesis operationalized by comparing successive immigrant cohorts. But the econometric literature on immigration shows that cohorts may not share common characteristics, in which case generalizations from the cross-section will not hold true. These caveats aside, the CPS provides a unique window into the patterns of union membership among America's new immigrant workforce. Background

This chapter stands at the confluence of two distinct literatures, one on immigration, the other on unionization, each of which is only peripherally aware of the other's existence. Labor matters play an important role in the historiography of American immigration (see Bodnar 1985; Barrett 1992), but even the historians have not yet generated a unified

Immigrant Workers and American labor

53

framework, integrating the insights from the study of immigration and the study of unionization. The contemporary literature is still more wanting. True, there are a small number of insightful case studies examining organizing among immigrant workers, of which the best known is Delgado's 1993 book; but these contributions tend to be nested more firmly within the immigration than the union literature. Labor economists have been moved to pursue an occasional inquiry; though this work evinces some methodological sophistication (for example, DeFreitas 1993), it provides a limited engagement with the broader intellectual issues at stake. In this section, we seek to distill lessons from both the immigration and the industrial relations literatures in an effort to specify the likely factors affecting unionization among today's immigrant workers.

Immigrants This chapter implicitly asks a simple question: Are immigrants more or less likely than other workers to have union jobs, other background factors controlled? The question has the appeal of simplicity; regrettably it is also simplistic, at least as concerns contemporary immigration to the United States. Earlier in the century, during the great wave of migration from southern and eastern Europe, the category "immigrant" captured a reasonably homogeneous experience. Of course, not all groups were the same; the Jews arriving from the Austro-Hungarian and Russian Empires differed in social characteristics and circumstances of migration from the Italians, the Slavs, and the Hungarians. But the Jews apart, and theirs was only a partial exception, most immigrants came with little skill; they moved into the bottom of the labor market, where they experienced a good deal of churning; often movement to the United States represented a temporary displacement on the route to return migration to the home country, if not the home community; and most immigrants were men, the advent of their wives and families representing a second-stage development signaling settlement. The foreign-born population that has emerged in the three and a half decades since the Hart-Celler Act marked the renewal of immigration is strikingly different. 3 Today's newcomers unquestionably include a sizable mass deeply reminiscent-in characteristics, circumstances of migration, and societal reaction-to the labor migrations of the 1880-1920 period. Indeed, immigrants from Mexico simultaneously comprise the least skilled and the largest single national origins group among today's new3. Portes and Rumbaut (1996) outline a typology of contemporary immigration in much greater detail; our discussion here draws on their influential formulation.

54

R 0 G E R W A l D I N G E R and C l AU D I A D E R - M A RT I R 0 S I AN

comers: as of the mid-1990s, 27 percent of all foreign-born persons residing in the United States were born in Mexico; persons with an elementary education or less comprised the single largest group of adults in this contingent. Like the Italians or the Poles, the Mexicans are involved in a pattern of circular migration, in which many, perhaps most, depart with the intention of returning home. Though the longevity of Mexican migration and its deeply implanted networks ensure that high proportions drop out of the circular streams and settle in the United States for good, the process of putting down roots remains a protracted affair. That extended transition is likely to exercise a powerful effect both on the social processes that connect workers to union jobs and on the preferences for union membership, for reasons discussed in greater detail below. There are other migrations, in addition to the Mexican movement to el norte, that are disproportionately dominated by proletarians who move in response to labor needs in the United States. Nonetheless, diversity in circumstances of migration as well as in socioeconomic characteristics represent the salient characteristics of today's newcomers. Ever since World War II, the U.S.-bound migration flows have contained an increasing number of refugees. The relative prominence of refugees reflects, in part, immigration's new role in U.S. foreign policy: until the end of the Cold War the doors were generally opened for refugees fleeing Communist regimes, while typically closed for all others, as in the Haitian or Salvadoran cases. Sociologically, the refugee population includes many besides those who officially qualify for refugee status, and one might expect the entire group to develop a propensity for both union jobs and union membership. As immigrants in a no-return situation, refugees typically have been quick to aspire to the standards enjoyed by the native population, a powerful reason why frustration in the search for upward mobility should produce a search for union jobs as well as pro-union sentiment. But their political history may push refugees in the opposite direction: as real or self-perceived victims of Communist regimes, many may be ideologically opposed to unions, despite strong instrumental considerations to do otherwise. Thanks to generous programs of government assistance, as well as the presence of former elites and home-country entrepreneurs, many refugee communities have developed ethnic economies of sizable employment potential, yet another reason why refugees may neither gravitate toward nor prefer union jobs. Though the refugee flow is impressive in both relative and absolute terms, what most clearly distinguishes today's situation from the past is the large proportion of highly educated persons among the immigrant

Immigrant Workers and American Labor

55

population. Highly-skilled immigrants have played a modest but significant role in immigration to the United States ever since the enactment of the Hart-Celler Act in 1965. Notwithstanding charges that America's immigrants are of "declining quality," the 1990 Census found that a college degree was as common among immigrants as among natives (one out of five). And among particular immigrant groups, the highly skilled are often present at levels well above the U.S. average, with the college graduate share ranging from 27 percent among Russians to 65 percent among Indians. A significant proportion of those immigrants who arrive with high skills discover that their premigration training yields no payoff in the United States. Those who undergo such "blocked mobility" seek to get ahead in different ways, often through entrepreneurship, which also takes them out of the supply of persons seeking union jobs. Not every group of highly skilled immigrants shows the same propensity for selfemployment; the Filipinos, for example, are highly educated but are particularly unlikely to work for themselves. More typical patterns, however, are the high self-employment rates among Koreans, Indians, Chinese, and Iranians, to cite the most notable cases. Whether as business owners or salaried professionals and managers, a good proportion of the recent arrivals begin not at the bottom but in the middle class or above (see Lopez and Feliciano, this volume). In contemporary Los Angeles, for example, certain coveted professional occupations have become immigrant concentrations: more than 35 percent of the pharmacists in the region are foreign-born, as are more than 25 percent of the dentists and over 20 percent of engineers, computer specialists, and physicians. As migration selectivity diminishes over time, most of the important U.S.-bound flows include persons from all parts of the occupational spectrum. Even so, significant interethnic differences in occupational composition persist, such that the proportion of highly educated persons or those working in upper white-collar professions distinguishes many of the flows from Asia from those that come from Mexico, Central America, or the Caribbean. The prevalence of professional, managerial, and entrepreneurial activities among immigrants implies that many can move ahead without the benefits that union jobs provide; that the size of this more highly skilled population varies considerably among the major immigrant streams reduces the likelihood that immigrant status as such will have a singular, unvarying effect on employment in a union job.

56

R0 G E R W A LD I N G E R and C LA U D I A D E R - M A RT I R0 S I A N

Work and its Rewards The relationship between the rewards of work and unionization preferences is well known. As Freeman and Medoff note, "The results of studies are unequivocal across very different samples. One finds that increased desire for unionization (expressed in union activity or votes for union) is, indeed, a likely outcome of worker dissatisfaction" (1984, 146). While one hardly expects it to be otherwise, knowing that more dissatisfied workers are more likely to prefer unions also begs the question, which must include both the causal nexus between satisfaction and union preference, on the one hand, and the factors that influence job satisfaction, on the other. In general, dissatisfaction with the bread-and-butter aspects of work provides the strongest push for workers to want union representation, especially among blue-collar workers. Though of lesser importance, job content also affects union preferences; not surprisingly, workers whose jobs possess more desirable features turn out to be less likely to voice support for unionism. These generalizations should cast immigrants as leading candidates for unionization while also generating a strong preference for union jobs, especially in light of unions' mitigating effects on undesirable job features. Immigrants, especially the least skilled, are more likely than natives to occupy jobs whose characteristics-low pay, unhealthy or dangerous working conditions, limited chances for promotion, and minimal job security-tend to be associated with higher levels of dissatisfaction. However, the conditions that generate dissatisfaction are surely evaluated in relative terms. While today's "bad jobs" might rank favorably when compared with the "bad jobs" of fifty years ago, what matters is how they contrast with today's average jobs. As Bakke put it in a classic article (albeit one composed in outdated vocabulary): The worker reacts favorably to union membership in proportion to the strength of his belief that this step will reduce his frustrations and anxieties and will further his opportunities relevant to the achievement of his standards of successful living (1967 [1945], 85). But if immigrants do not use the same yardstick as natives, then the factors making for frustration are likely to be quite different. For many immigrants, the relevant standard is defined by significantly inferior conditions back home, not those that prevail in the United States. From that perspective, employment in a sewing factory or as a janitor cleaning office buildings not only ranks higher in the status hierarchy in which the immigrant grew up, but provides material rewards the likes of which the

Immigrant Workers and American Labor

57

newcomer never knew before. While the comparative frame changes over time, as exposure to the United States and its consumption standards pushes the immigrant's normative expectations higher, other factorsmost notably, continued contact with and return travel to the home society-keep the older normative pattern in place. Not every group is likely to experience change in normative expectations at the same rate. Those most committed to long-term settlement in the United States are most likely to experience rapid convergence with U.S.-based norms. But this is precisely why the most proletarian migrants may be least likely to experience their conditions as dissatisfYing. Unlike their more skilled counterparts, the labor migrants begin as target earners, concerned with short-term rather than long-term rewards and assessing the adequacy of those rewards relative to some consumption objective located back home. Many of these initially temporary immigrants eventually drop out of the circular migration stream, their standards shifting as they settle down. 4 Even so, the phasing out of the back-and-forth flow is often an extended process, and the majority of those with at least one migratory experience probably never drop out-which means that at any point in time a large proportion of the most proletarianized immigrants see themselves as more likely to return home than to settle down. "Demographic" Characteristics

So unhappy workers are likely to want unions. Some groups of workersAfrican Americans most definitely, women quite possibly-are both unhappier and more union-prone than the rest. The pro-union inclinations of African American workers is in fact one of the few matters on which almost all researchers who have studied the matter can agree; they are significantly more likely to want union representation and to vote for union representation, when given the choice, than whites. Furthermore, that relationship holds after controlling for background characteristics, including the much higher likelihood that African Americans will fill jobs at the lower end of the wage distribution, a characteristic strongly conducive to a union propensity regardless of race or ethnicity. But the industrial relations literature is a good deal less certain as to how these intergroup differences in unionization propensities should be interpreted. As Farber and Saks suggest (1980), it may simply come down to differences in "taste": group differences in preference for unions may be exogenously determined. Alternatively, groups may make systematically distinctive evaluations of individual needs, on the one hand, and of 4. For elaborations of this theme, see Piore 1979 and Massey eta!. 1987.

58

R 0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R 0 S I A N

the rewards provided by employers and unions on the other. In particular, intergroup perceptions of the likelihood of employment discrimination and the potential for redress offered by unions could significantly affect both union propensities and preferences for union jobs. Indeed, Mrican Americans discover that the wage standardizing and grievance procedures found in unionized establishments significantly reduce discrimination (Ashenfelter 1972). For that reason, it seems reasonable to conclude that "the 'nonwhite effect' appears to be a 'real' demographic effect in the sense that research has not yet developed attitudinal measures which would better account for differential unionism preferences" (Fiorito, Gallagher, and Greer 1986, 279). This formulation implicitly assumes that such traits as race and ethnicity are real, existing characteristics of the individual, such that one's own and others' understanding of membership status are clear, consistent, and symmetrical. But this description fits awkwardly with the situation of immigrant workers, for whom membership in an ethnic group is not imported, but rather a phenomenon that emerges out of the migration experience itself. Immigrants do not arrive as "ethnics" but rather become ethnics. On the one hand, categorization as "other"-by dominants, competitors, and members of one's putative group-alters selfunderstanding; and on the other hand, shared experience with similarly categorized others imparts a sense of solidarity and an awareness of common interest. Most important, stigmatization and exclusion produce areactive ethnicity, in which membership in a group is defined by virtue of opposition to dominants. Assimilation The concept of assimilation provides yet another key for understanding how immigrants may both secure union jobs and also come to change their views regarding the advantages of unionization. Of course, there are few more contested concepts in the social sciences than assimilation, and now is not the place and time to enter into that debate. The conventional definition will do just fine for the purposes of this chapter: assimilation involves the process whereby immigrants become increasingly like natives on a number of dimensions, moving up the economic ladder as they gain in skills and language ability, while also absorbing the values, orientations, and preferences held by the native-born population. In this sense assimilation might be expected both to generate the skills that will allow immigrants to obtain relatively scarce union jobs and also foster the preferences that would motivate immigrants to either seek out those jobs or

Immigrant Workers and American Labor

59

unionize their unorganized employer, at rates more or less comparable to counterparts in the native population. 5 Whether this hypothesis bears true will partly depend on how immigrants evaluate the costs and benefits of union membership. Since unions tend to reduce wage differences, the econometric literature emphasizes the importance of workers' place in the wage hierarchy, with those below the median wage more likely to benefit from the union wage effect and therefore more likely to seek union jobs, and those above the median wage influenced in exactly the opposite way (Abowd and Farber 1982). At the very least, the heterogeneity of today's immigrant population ensures that the economic appeal of union membership will vary greatly. To the extent that a substantial portion of today's newcomers move immediately or quickly into the middle class, their assimilation trajectories are unlikely to lead them toward union jobs or make those jobs attractive. Not every immigrant group, moreover, will assimilate the expectations and orientations of the native population in identical rate or fashion, regardless of how proletarian its background. Workers from groups in which refugees predominate are likely to experience a quicker commitment to life in the host society, accelerating the rate at which they aspire to the average standards of compensation and therefore come to appreciate the redistributional union wage effect. By contrast, workers emanating from groups with a history of circular or temporary migration will be more likely to retain a dual frame of reference, which in turn reduces the impetus to seek out union jobs. Not only are immigrants thus likely to assimilate economic expectations and orientations at varying rates, the social nature of the processes of migration and incorporation makes for divergent paths up from the bottom. As Partes and Rumbaut argue, the fate of new arrivals largely depends "on the kind of community created by their co-nationals" (1996, 84). Labor migrants enjoy easy access to jobs at the bottom of the labor market, thanks to a dense web of social networks linking veterans and newcomers. However, those same connections funnel labor migrants into a narrow set of occupations and industries from which exit is difficult, in large measure because dependence on ethnic networks chokes off the flow of new information, constraining diffusion and the search for new opportunities. Similar influences affect members of more entrepreneur5. For a prominent recent example of a largely unself-conscious application of assimilation, see the National Research Council's report, The New Americans (Smith and Edmonston 1996). For the controversies surrounding the use and nature of the concept, see the essays in the special issue of the International Migration Revzew 31: 4 ( 1997).

60

R0 G E R W A LD I N G E R and C LA U D I A D E R - M A RT I R 0 S I A N

ial streams while pushing them in a different direction: in this case, the business success of co-ethnics gives rise to a distinctive motivational structure, breeding a community-wide orientation toward small business and encouraging the acquisition of skills within a stable, commonly accepted framework. Regardless of the precise modality, the social structure of the immigrant community has an enduring effect on the trajectories that immigrants follow, reducing the likelihood that otherwise comparable immigrants will move into union jobs at a similar pace.

Summary Thus the process by which immigrants develop a "demand for union services," to borrow the economists' language, is likely to take a variety of forms, proceeding along not one but several timetables. For all immigrants, time is likely to be the most decisive consideration, as both interest in the benefits obtained through unionization and the ability to obtain union jobs increase with time spent in the United States. In this respect, time points to the importance of assimilation: because standards change with settlement, the longer immigrants live in the United States the more likely their frame of reference will shift from the conditions prevailing in the country of origin to those prevailing in their new home. However, time does not yield a uniform effect. Workers emanating from groups with a history of circular or temporary migration may be more likely to assimilate in a delayed fashion. Workers from groups dominated by refugee flows are likely to experience a quicker commitment to life in the host society, which in turn will accelerate the impact of time." Furthermore, time generates new self-understandings, and not simply because immigrants come to view themselves in terms consistent with the categories of the U.S. system of race and ethnic relations. As immigrants become ethnics, their perception of fairness in the employment system evolves, as does their view of the need for redress through unions. Reactive ethnicity is likely to exercise a powerful effect on Mexican immigrants: though preferences for unionization will be depressed, relative to other immigrants, at early stages of the migration, they should shift more sharply upward in the later stages of settlement. On the other hand, the social structure of Mexican migration is likely to impede access to existing union jobs, even relative to other immigrants of similar socioeconomic background. The socioeconomic diversity of the immigrant population further complicates the impact of time. For the large middle-class contingent 6. Unfortunately, our sample size precludes an adequate test of this latter hypothesis.

Immigrant Workers and American Labor

61

among today's immigrants, occupational position will either preclude unionization-as among those who work as managers-or reduce its attractiveness or feasibility-as among the many immigrants who are employed in higher or even low white-collar positions. The importance of socioeconomic diversity suggests that controls for demographic characteristics and for industry or occupation will reduce the effects of immigrant status and time.

Data and Variables This section analyzes data drawn from a four-year (1994-97) combined Current Population Survey (CPS) for the month of March. 7 The CPS inquires into union membership among only a relatively small portion of eligible respondents, and we have restricted the sample further to encompass only those aged 20-65 and who were also employed as wage and salary workers. The analysis, therefore, excludes the self-employed. The analysis is based on a sample of almost 32,000 persons, 2,998 of whom are foreign-born. The limited size of the foreign-born sample makes meaningful national-origin disaggregations for most immigrants problematic. Consequently we have merged data from several questions to recode the sample into five mutually exclusive "ethno-racial" groups: whites, blacks, Mexicans, other Hispanics, and others, the latter a largely Asian grouping. We also use controls for nativity and year of immigration to distinguish native from foreign-born members of these groupings. As additional controls, we use data on geographical location, gender, education, age, and industry of employment. 8 7. The Current Population Survey is a monthly survey of a national probability sample of approximately sixty thousand households. In light of the limited size of the two popu· lations of interest to us-immigrants on the one hand and union members on the otherwe have sought to increase the size of these target populations by merging the March CPS samples from 1994 through 1997. However, the nature of the CPS precludes utilization of each year's full CPS sample. The CPS retains respondents during a two-year period, inter· viewing individuals for four consecutive months, dropping them from the sample for the next eight months, and then re-interviewing them for another four consecutive months, after which time they are dropped from the sample completely. Consequently, half of the persons interviewed in any given month reappear in the following year's sample in the same month. To avoid duplicate cases, we have retained non overlapping halves of the 1994, 1995, and 1996 samples, and included the entire 1997 sample. This procedure almost triples the size of the sample. 8. We have recoded industry of employment into six mutually exclusive categories, separating the public sector from all private sector industries, and recoding private sector industries into five categories: construction I manufacturing; agriculture, mining, and forestry; transportation, communications, and utilities; wholesale I retail trade; and finance, insurance, real estate and other services.

62

R0 G E R W A LD I N G E R and C LA U D I A D E R - M A RT I R0 S I AN

Characteristics The immigrants in our sample differ from natives along a series of dimensions, as table 2.1 shows. First of all, natives are more likely to hold union jobs: just over 18 percent of native-born workers are unionized, as opposed to just over 15 percent for the foreign-born. Differences in age and labor force experience are negligible, in part because our sample is limited to workers aged 20-65. Immigrants are more heavily male than natives, which makes the difference in unionization rates even more striking, as generally male workers in the U.S. are more unionized than female workers. Still more impressive are native I immigrant differences in ethnic characteristics. The native-born population in our sample remains overwhelmingly white. By contrast, whites comprise less than a quarter of the foreign-born; Hispanics, divided here between Mexicans and other Hispanics, account for the largest group of immigrants, followed by "Asians I Others." By historical standards, relatively high skill levels distinguish today's immigrant population; but a smaller proportion of immigrants have gone beyond the high school degree than is true among the native-born. A much greater difference is found at the lower end of the spectrum: immigrants are disproportionately concentrated among persons who have not obtained the high school degree- a very small group among the native-born. Though found in all major industrial sectors, immigrants evince a different sectoral distribution than natives. Construction and manufacturing constitute a sector of immigrant overconcentration; the public sector, not surprisingly, has relatively low levels of foreign-born employment. We also note the relatively recent origins of the immigrant workforce, of whom just under 60 percent arrived in the United States after 1980. Even this figure probably underestimates new arrivals' share of the immigrant population, since our merged (1994-97) sample is such that it does not fully capture immigration throughout the 1990s. Geography offers a final axis of differentiation. The geographic distribution of immigrants and natives is almost perfectly asymmetrical, the six states containing 65 percent of the foreign-born being home to less than 30 percent of all natives. California, with 26 percent of the foreignborn workers, contains only 6 percent of all native-born adults of working age.

Immigrant Workers and American Labor

63

Table 2.1. Characteristics by nativity, wage and salary workers aged

20-65.

Natives Holding union jobs Mean age Mean labor market experience Male Education Less than high school High school diploma More than high school State California New York New Jersey Illinois Florida Texas Other states Race White Black Mexican Other Hispanic Asian/ other Industry Agriculture I mining I forestry Construction/ manufacturing Trans. I comm. I utilities Wholesale I retail trade Fin. I ins. I real est. I other services Public admin. I public sector Period of Immigration Before 1980 1980-1989 1990-1997 U.S. Citizen

(n = 27,703)

Immigrants (n = 2,998)

18.1% 40.3 20.9 50.0%

15.1% 39.8 21.6 56.0%

7.6% 33.7% 58.7%

28.0% 22.7% 49.4%

5.8% 5.5% 3.6% 4.1% 4.1% 4.7% 72.3%

25.9% 12.8% 8.0% 4.8% 8.1% 5.5% 34.8%

83.2% 9.8% 2.7% 1.8% 2.6%

24.3% 6.3% 23.2% 19.1% 27.2%

1.7% 21.7% 6.4% 18.5% 32.9% 18.9%

3.6% 26.9% 4.3% 18.5% 36.2% 10.5% 42.4% 36.1% 21.6% 36.1%

64

R 0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R 0 S I A N

But it is not simply that immigrant and native workers live in different regions of the country; there are geographic variations within the immigrant population as well. Those many immigrants who move to California differ significantly from their counterparts who settle elsewhere, as table 2.2 shows. Disparities in native-immigrant unionization rates stand at the top of the list: natives are unionized at almost twice the rate of immigrants in California, whereas elsewhere the difference is on the order of less than 20 percent. Educational characteristics also make California distinct; its native-born workers are far better educated than those elsewhere, but its immigrants are more likely to be concentrated among the very least schooled, making for an extraordinary large educational gap between natives and immigrants. More than half of California's immigrants are Hispanic, almost three-quarters of whom are Mexican; elsewhere, Hispanics comprise less than 40 percent of the foreign-born workforce. And California's immigrants are longer settled than those residing elsewhere in the United States, a characteristic reflecting California's importance as a prime immigrant destination ever since the late 1960s. Thus California stands out from the rest of the nation with respect to the social characteristics of its immigrants, and with respect to the size and nature of differences between immigrants and natives. Not surprisingly, those differences are associated with distinctive patterns in immigrant/native unionization rates, as tables 2.3 and 2.4 show. Overall, California still ranks as a state with relatively high unionization rates. Native-born workers are just a little less likely to have union jobs than their counterparts in New York or New Jersey, but far more so than in such immigrant states as Texas or Florida. Immigrant Californians are far less unionized than their New Jersey or New York counterparts, but compare favorably with foreign-born workers in Texas or Florida. Nonetheless, a sizable foreign-native gap in unionization rates distinguishes Californians within almost every social category, residents of San Francisco excepted. Whether men or women, better educated or less educated, Mexicans or Asians, manufacturing workers or workers in trade, immigrant Californians are far less likely to hold union jobs than their native-born counterparts. Immigrants in the other forty-nine states are also less likely to hold unionjobs than comparable native-born workers, but the gap is modest, especially by comparison to California. Moreover, unionization rates among recent immigrants to California are highly depressed.

Immigrant Workers and American labor

65

Table 2.2. Characteristics by nativity, wage and salary workers aged

20-65, California vs. other 49 states.

Other 49 states

California

Natives Immigrants Natives (n 1,614) (n 777) (n 26,089)

Immigrants (n =2,221)

24.7% 39.8

13.4% 38.3

17.7% 40.4

15.8% 40.3

19.9 51.0%

20.6 59.2%

21.0 49.9%

21.9 54.8%

5.1% 24.9% 70.4%

33.1% 21.1% 45.8%

7.7% 34.3% 58.0%

26.2% 23.2% 50.6%

51.4% 20.9% 25.2% 2.5%

68.7% 16.6% 13.5% 1.2%

71.7% 7.5% 13.6% 2.6% 4.7%

13.6% 1.9% 39.8% 14.4% 30.2%

83.9% 10.0% 2.0% 1.7% 2.5%

27.9% 7.8% 17.4% 20.7% 26.1%

0.9%

3.9%

1.8%

3.5%

17.9% 7.5% 18.5%

30.2% 4.0% 18.4%

21.9% 6.3% 18.5%

25.7% 4.4% 18.5%

35.4%

33.3%

32.8%

37.2%

19.8%

10.2%

18.8%

10.6%

=

Holding union jobs Mean age Mean labor market experience Male Education Less than high school High school diploma More than high school Location Los Angeles San Francisco Other metropolitan Non-metro Race White Black Mexican Other Hispanic Asian/ other Industry Agriculture I mining I forestry Construction I manufacturing Trans. I comm. I utilities Wholesale I retail trade Fin. I ins. I real est. I other services Public admin. / public sector Period of immigration Before 1980 1980-1989 1990-1997 U.S. citizen

=

41.6% 42.5% 16.0% 30.0%

=

42.6% 33.9% 23.5% 38.3%

66

ROGER WALDINGER and CLAUDIA DER-MARTIROSIAN

Table 2.3. Proportion ofwage and salary workers aged 20-65 holding union jobs, United States (percentage). Natives (n

Gender Male Female Education Less than high school High school diploma More than high school State California New York New jersey Illinois Florida Texas Other states Race White Black Mexican Other Hispanic Asian I other Industry: Agriculture I mining I forestry Construction I manufacturing Trans. I comm. I utilities Wholesale I retail trade Fin. I ins. I real est. I other services Public admin. I public sector Period of immigration Before 1980 1980-1989 1990-1997 U.S. Citizen Yes No

=27,703)

Immigrants (n 2,998)

=

21.0% 15.2%

15.7% 14.4%

14.8% 19.1% 24.9%

14.7% 17.4% 16.9%

24.7% 29.5% 23.9% 21.5% 12.3% 9.1% 16.6%

11.7% 26.4% 25.2% 15.9% 5.7% 7.8% 12.8%

17.2% 28.1% 24.7% 23.6% 25.1%

16.2% 23.4% 10.3% 16.2% 14.4%

9.6% 20.0% 32.1% 6.5% 5.9% 44.7%

1.9% 15.1% 24.8% 6.3% 11.4% 44.1% 25.0% 12.9% 9.4% 20.50% 12.10%

Immigrant Workers and Ameriran labor

67

Table 2.4. Proportion of wage and salary workers aged 20-65 holding union jobs, California vs. other 49 states (percentage). California Natives (n = 1,614)

Gender Male Female Education Less than high school High school diploma More than high school Location Los Angeles San Francisco Other metropolitan Non-metro Race White Black Mexican Other Hispanic Asian I other Industry: Agriculture I mining I forestry Construction I manufacturing Trans. I comm. I utilities Wholesale I retail trade Fin. I ins. I real est. I other services Public admin. I public sector Period of immigration Before 1980 1980-1989 1990-1997 U.S. Citizen Yes No

Other 49 states

hnmigrants Natives (n= 777) (n = 26,089)

Immigrants (n = 2,221)

28.4% 20.9%

14.6% 11.7%

20.6% 14.8%

16.2% 15.3%

16.9% 25.8% 24.9%

9.7% 11.6% 16.9%

14.7% 18.8% 17.5%

16.8% 19.2% 13.6%

26.4% 24.9% 22.4% 12.5%

11.4% 20.9% 14.3% 11.1%

24.0% 28.1% 25.6% 19.1% 30.7%

17.9% 26.7% 12.3% 15.2% 11.1%

16.80% 24.50% 13.90% 24.00% 20.40%

15.90% 23.10% 8.80% 18.70% 15.70%

13.3%

0.0%

9.5%

2.6%

18.0% 35.5% 13.4%

8.9% 19.4% 5.6%

20.1% 31.9% 6.1%

17.7% 26.5% 6.6%

10.0%

10.0%

5.6%

11.9%

64.1%

54.4%

43.4%

40.7%

21.4% 8.8% 4.8%

19.5% 14.6% 10.5%

24.90% 8.50%

19.3% 13.6%

68

R 0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R 0 S I A N

Multivariate Analysis Multivariate analysis can provide a more rigorous assessment of the factors associated with the probability of holding a union job. We seek to determine the likelihood of being a union member; as membership is a dichotomous category, we use logistic regression to predict the probability of membership status. 9 Because the coefficients produced by logistic regression are difficult to interpret, we exponentiate them to generate odds ratios. An odds ratio less than one indicates a negative relationship between a given independent variable and the probability of being unionized; an odds ratio greater than one points to a positive relationship. To enhance readability and facilitate intuitive understanding, our discussion below refers to graphical representations of the odds ratios. Detailed results appear in the chapter appendices. We will first examine patterns for the entire United States, and then review the findings for California. We begin by looking at the effects of time, asking whether immigrant/native differences in unionization diminish as length of settlement in the United States increases. As figure 2.1 shows, immigrants in the more recent cohorts are indeed significantly less likely to be unionized than native workers; among immigrants of the longest tenure, however, foreign birth proves to be positively associated with union membership. The overall pattern persists after controlling for ethnicity, with other Hispanics, blacks, and others significantly more likely to be union members than whites, and Mexicans significantly less likely to be union members. Because there are strong effects linking ethnicity and union membership, controls for ethnicity slightly mute the impact of time among the most established of the foreign-born. Applying controls for demographic characteristics eliminates the union propensity among the longest-settled immigrants, while leaving the effects of migration in either the 1980s or 1990s virtually unchanged. Insertion of the industry dummies alters the association between immigrant status and union membership, though the effect is inconsistent: older immigrants are now 9. When analyzing categorical dependent variables, linear regression is not an appropriate method. Linear regression assumes continuous dependent variables. In analyzing our dichotomous dependent variable (union membership), we used logistic (or logit} regression. Logistic regression predicts the probability that an event occurs. Unlike linear regression, logit (or logistic) regression refers to models with a logit as the left-hand-side variable. A logit is the log of odds that an event occurs. Thus the estimated (logit) coefficients are reported in terms oflog odds. The interpretation oflog odds is rather difficult. Instead of log odds, odds allow a more intuitive interpretation, especially with dichotomous predictors. For dummy predictors, the odds ratio equals the antilogarithm of the logit coefficient. In this chapter, we present odds ratios for all predictors.

69

Immigrant Workers and American Labor 1.4 1.2

0.8

0.6

0.4 0.2

0 Arriving 1979 and before • No controls

0 Race/ethnicity

Arriving 1980s • Demographics

Arriving 1990s

0 Industry

• Location

Figure 2.1. Odds of holding union job, United States: effect of immigrant cohort before and after controls.

significantly more likely than natives to hold union jobs; immigrants of the 1980s no longer differ from native counterparts; and though immigrants ofthe 1990s remain less likely than natives to hold union jobs, the effect is no longer as strong. Still, the most recent immigrants are only about half as likely as natives to be union members, other relevant background factors being controlled (see appendix A). Location Immigrants tend to converge on just a handful of states; ironically, those same immigrant-receiving states differ significantly in the industrial relations environment, as we've already noted. Compared to Californians, workers in New York are about 50 percent more likely and those in Florida over 50 percent less likely to hold union jobs. More important, adding variables for location to the controls applied earlier alters the effects and significance associated with migration cohorts: newcomers of the 1980s or the 1990s again become significantly less likely to be union members; older migrants are now no more likely than natives to hold union jobs.

70

R0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R0 S I A N

Time I Ethnicity Interactions Unionization rates among immigrants thus increase over time; immigrants with eighteen years or more of residence in the United States are just as likely to be union members as their native-born counterparts. More recent immigrants, however, show a strong union membership gap, a divergence that bears little relationship to background demographic factors and is only mildly affected by sectoral location. But time may not work in quite the same way for all groups. In particular, labor migrants involved in circular movements from home to host society and back again may be slowest to adjust their standards of evaluation to the circumstances that prevail in the host society. Thus we hypothesize that Mexicans, as the quintessential labor migration group, will display a distinctive pattern, in that recency of arrival will have an even more negative effect on unionization status than among all other immigrants. Indeed, as figure 2.2 shows, Mexican immigrants of the 1980s and of the 1990s are significantly less likely to be union members than members of all other immigrant groups. While the divergence is greatest among the immigrants of the 1990s, the cohort effect appears quite robust; ap-

Mexicans

Whites

0 Arriving 1979 and before

Other Hispanics Arriving 1980s

Others/Asians

• Arriving 1990s

Figure 2.2. Odds of holding union job, United States: immigrant cohorts

by ethnicity (background factors controlled).

Immigrant Workers and American Labor

71

plication of demographic and industrial controls shows that recent Mexican immigrants are less likely to be union members than their demographic counterparts working in the same industries.

The California Factor Multivariate analysis confirms the initial comparison between the United States and California: immigrants in California are less likely to hold union jobs, as figure 2.3 shows. But unlike the pattern in the United States as a whole, immigrant Californians are significantly less likely to hold union jobs than native-born workers regardless of cohort, with the unionization gap especially wide among more recent arrivals. Sequential controls for ethnicity, demographics, sector, and location slightly diminish the difference between natives and immigrants in each cohort, but without altering the essential pattern: for immigrant Californians, a union job remains, if not out of reach, then awfully hard to find (see appendix B) .

0.9 0.8 0.7

0.6 0.5 0.4

0.3

0.2 0.1

0 Arriving 1979 and before • No controls

0 Race/ethnicity

Arriving 1980s Demographics

Arriving 1990s

0 Industry

• Location

Fig;ure 2.3. Odds of holding union job, California: effect of immigrant cohort before and after controls.

72

R 0 G E R W A l D I N G E R and C l A U D I A D E R - M A RT I R0 S I A N

Propensity or Social Process? Thus our analysis shows that immigrants are less likely than natives to hold union jobs for reasons that have relatively little to do with the social factors that otherwise distinguish immigrants from natives. But if background characteristics don't account for the disparity, what does? In the current environment, group differences in unionization rates are likely to be closely related to group differences in access to union jobs; as we've suggested, the process of immigrant incorporation systematically reduces the probability of moving into those sectors or organizations where union jobs are most likely to be found. On the other hand, circumstances of migration and settlement generate distinctive economic expectations and job preferences that can affect access to the types of positions that tend to be covered by union contracts. So can one reduce intergroup differences in unionization rates to employment patterns? Or do inter-group propensities for unionization play any role in the process? Our data preclude any definitive assessment of immigrant propensities to unionization, but they do allow for an indirect test of the hypothesis that the circumstances of immigrant incorporation and settlement

1.6

Arriving 1979 and before

a

No controls

Arriving 1980s

Arriving 1990s

• With controls

Figure 2.4. Odds of union membership, United States: effect of immigrant cohort before and after controls.

Immigrant Workers and American Labor

73

may reduce the preference for unionization. Though getting a union job may be a matter of employment patterns, persons who hold such jobs choose whether to become union members or not. Of those persons in our sample who hold union jobs, as we've defined them in this chapter, 10.5 percent are not union members. As figure 2.4 shows, although immigrants who arrived before 1980 are more likely than their native-born counterparts to be union members, if covered by a union contract, those immigrants of the 1980s and 1990s who hold union jobs are significantly less likely to be union members than native-born unionized workers, a disparity that holds after controlling for education.

Conclusion In this essay, we plumb a new data source to assess the factors associated with unionization among America's new immigrant workers. The effort yields a sobering conclusion: however much benefit they might derive from unionization, immigrant workers have difficulty finding union jobs. To be sure, the picture is not uniformly negative. Over time, unionization rates rise among immigrants, so that among the more settled of the arrivals, unionization rates are somewhat higher than among their nativeborn counterparts. However, the more recent cohorts are distinctly less likely to be union members than their native-born counterparts, a disparity that persists after controlling for a broad range of background factors. Furthermore, the difference between natives and immigrants is even greater among the largest, and most uniformly proletarian of contemporary immigrant groups, namely the Mexicans, among whom recent arrivals are especially unlikely to be union members. Union jobs are most elusive in California, home to one out of every four immigrant workers, where even the most settled immigrant workers are unlikely to hold a union job. Our analysis is pitched in the language of variables, but it's always important to recall the distribution of populations among the values on those variables. From this perspective, our findings regarding unionization rates among the more recent cohorts would be of considerably reduced import had immigration to America subsided; were most immigrants of established vintage, recentness of arrival would have little net effect on the overall native I immigrant unionization gap. Similarly, if more immigrants lived in New York than California, recency would have a much milder overall effect, and immigrant unionization rates would soon surpass those of the native-born. However, the reality is rather different.

74

R 0 G E R W A L D I N G E R and C LA U D I A D E R - M A RT I R 0 S I AN

The great majority of immigrant workers have moved to the United States since 1980, with two-and-a-half times as many in California as in New York. Working-class immigrants have other characteristics-such as low levels of education and concentrations in industries of low union densities-which further reduce access to union jobs. This paper can shed light only on the factors associated with access to existing union jobs, and that is a matter oflittle import to the future of organized labor. Immigrant workers may well find it difficult to connect with those jobs located in the shrinking union sector, but the factors influencing their prospective behavior may prove more crucial. Indeed, the central finding of this chapter is consistent with other research documenting the troubles that immigrants encounter in their search for progress in America. Today's newcomers thus have ample reason for turning to unions to voice their discontent; they are also quite capable of doing so, as the other chapters in this volume show. Nonetheless, the factors that make union jobs elusive may weaken the appeal of unionization itself. Immigrant modes of incorporation generate a logic of their own; such logic will not always breed a preference for unionization, job problems notwithstanding. Likewise, continuing homecountry ties generate a dual frame of reference, which in turn blunts the impact of those frustrations encountered in the United States. And as we noted above, the limited evidence available does suggest that recent immigrants are less likely than comparable natives to choose union membership, when offered the possibility. Thus immigration is unlikely to add a silver lining to the dark clouds facing American labor. Granted, labor has been misled in recent times; it~ new leadership-should it last-provides some reason for hope. Yet the obstacles it confronts are so deep-seated that even a greatly improved leadership equipped with superior strategy may not suffice. As we've argued, the characteristics of the new immigrant workforce-its size, composition, and most important its recent arrival-strongly suggest that immigration adds to, rather than subtracts from, the many difficulties facing organized labor. While we admire the optimism that impels union organizers in the face of these odds, the argument and analysis in this chapter warrant pessimism, which seems very much the order of the day.

Appendix A

Logistic Regression Results Predicting Holding Union jobs, United States, 1994-97 (odds ratios for figure 2.1 in bold) Model A

Odds ratio

Std. error

z

P>lzl

bef1980 1980-89 1990-97

1.201839 0.7007825 0.5344614

0.0794123 0.0661929 0.0728039

2.782 -3.764 -4.599

0.005 0 0

1.055851 0.5823478 0.409229

1.368012 0.8433037 0.6980175

1.115299 0.6441107 0.4994239 1.599969 0.8522537 1.475955 1.16258

0.0824009 0.0659292 0.0709873 0.0753601 0.0758867 0.1317466 0.0948038

1.477 -4.298 -4.885 9.978 -1.795 4.361 1.847

0.14 0 0 0 0.073 0 0.065

0.9649452 0.5270285 0.3779905 1.458878 0.7157738 1.239062 0.9908562

1.289081 0.7872033 0.6598691 1.754704 1.014757 1.758138 1.364065

0.9692626 0.6497235 0.5621713 1.777984 1.051407 1.742445 1.262371 1.066229 0.6570613 1.027302

0.0733386 0.0674068 0.0809544 0.0859411 0.0967923 0.1593462 0.1046999 0.012217 0.0206463 0.0134818

-0.413 -4.156 -4 1.906 0.545 6.072 2.809 5.597 -13.366 2.052

0.68 0 0 0 0.586 0 0.005 0 0 0.04

0.8356721 0.5301747 0.4239289 1.617275 0.8778278 1.456521 1.072974 1.042551 0.6178162 1.001215

1.124209 0.7962293 0.7454942 1.954662 1.259309 2.084497 1.4852 1.090445 0.6987993 1.054068

[95% conf. interval]

ModelE bef1980 1980-89 1990-97 blacks mexicans other hispanics other Modele bef1980 1980-89 1990-97 blacks mexicans other hispanics other age female work experience

75

APPENDIX A

76

exper squared less than HS more than HS

Odds ratio

Std. error

z

P>lzl

[95% conf. interval]

0.9986955 0.8538642 0.8569842

0.0001151 0.0622357 0.0406179

-11.323 -2.168 -3.256

0 0.03 0.001

0.9984698 0.7401968 0.7809603

0.9989212 0.9849869 0.9404088

1.230827 0.9605481 0.7440377 1.564817 0.8654121 1.515685 1.020046 1.001278 0.6951388 1.07014 0.9989847 0.7576113 0.8335047 0.1108646 0.2782648 0.520164 0.0885022 0.0904891

0.1000538 0.1047883 0.1113939 0.0825558 0.0846433 0.1493756 0.0902508 0.0125183 0.024655 0.0152119 0.0001227 0.0594396 0.0425263 0.017489 0.0126843 0.0312589 0.0056713 0.0045899

2.555 -0.369 -1.975 8.487 -1.478 4.22 0.224 0.102 -10.253 4.769 -8.268 -3.538 -3.569 -13.942 -28.062 -10.876 -37.839 -47.365

0.011 0.712 0.048 0 0.139 0 0.823 0.919 0 0 0 0 0 0 0 0 0 0

1.049549 0.775639 0.5548262 1.411095 0.7144462 1.249454 0.857646 0.9770408 0.6484573 1.040737 0.9987442 0.649627 0.7541867 0.0813795 0.2544821 0.4623682 0.0780564 0.0819258

1.443414 1.189539 0.9977756 1.735285 1.048278 1.838644 1.213198 1.026116 0.7451809 1.100374 0.9992253 0.8835453 0.9211646 0.1510328 0.30427 0.5851843 0.1003459 0.0999475

1.06378 0.76843 0.6422665 1.592599 1.012552 1.39744 1.013749 0.993819 0.6935151 1.07867 0.9989659 0.7495692 0.8409902 0.1168272 0.2711065 0.5037195 0.0857332

0.0886573 0.0859561 0.0969293 0.0850411 0.104884 0.1412378 0.0910982 0.0125426 0.0248086 0.0154765 0.0001239 0.0594657 0.0433575 0.0184988 0.0125008 0.0307303 0.0055388

0.742 -2.355 -2.934 8.715 0.12 3.311 0.152 -0.491 -10.231 5.278 -8.339 -3.633 -3.359 -13.56 -28.307 -11.24 -38.024

0.458 0.019 0.003 0 0.904 0.001 0.879 0.623 0 0 0 0 0.001 0 0 0 0

0.9034648 0.6171482 0.4778086 1.434348 0.8265068 1.146313 0.8500396 0.9695376 0.6465567 1.048759 0.998723 0.6416279 0.7601634 0.0856568 0.2476799 0.4469508 0.0755366

1.252543 0.9567955 0.8633294 1.768311 1.240475 1.703583 1.208987 1.018709 0.7438841 1.109434 0.9992089 0.8756696 0.9304112 0.1593403 0.2967489 0.5676985 0.0973062

ModelD befl980 1980-89 1990-97 blacks mexicans other hispanics other age female work experience expcr squared less than HS more than HS agr I min I fores constlmfg

TCU

trade FIRE I services ModelE befl980 1980-89 1990-97 blacks mexicans other hispanics other age female work experience exper squared less than HS more than HS agr I min I fores constlmfg

TCU

trade

Immigrant Workers and American Labor

FIRE I services NY

NJ

IL FL

TX other states

77

Odds ratio

Std. error

z

P>lzl

0.0851743 1.459146 1.14397 0.8705907 0.3791937 0.6213894 0.2715013

0.0043843 0.1354191 0.1187679 0.0911953 0.0446571 0.0471358 0.0324886

-47.85 4.071 1.296 -1.323 -8.234 -6.272 -10.896

0 0 0.195 0.186 0 0 0

[95% conf. interval] 0.0770004 1.216469 0.9333455 0.7090057 0.3010345 0.5355444 0.214741

0.0942157 1.750235 1.402126 1.069002 0.4776457 0.7209949 0.3432646

Appendix B

Logistic Regression Results Predicting Holding Union jobs, California, 1994-97 (odds ratios for figure 2.3 in bold) Model A

Odds ratio

Std. error

z

P>lzl

bef1980 1980-89 1990-97

0.7644336 0.2808074 0.1503715

0.1109301 0.0569236 0.0635172

-1.851 -6.265 -4.485

0.064 0 0

0.5751994 0.1887374 0.0657077

1.015924 0.4177909 0.3441239

0.7549722 0.2756519 0.1477767 1.308275 1.039622 1.083548 1.047406

0.1257016 0.0610945 0.0636305 0.2667965 0.1522656 0.2698437 0.203712

-1.688 -5.814 -4.441 1.318 0.265 0.322 0.238

0.091 0 0 0.188 0.791 0.747 0.812

0.5447627 0.1785268 0.0635464 0.8772306 0.7802008 0.6650698 0.7154235

1.046296 0.4256164 0.3436538 1.951123 1.385302 1.765342 1.53344

0.6955648 0.3173513 0.1984953 1.39704 1.452269 1.306141 1.025638 1.029152 0.6566181

0.122331 0.0720513 0.0866765 0.2911591 0.2332618 0.3319918 0.2062967 0.0357719 0.0691256

-2.064 -5.055 -3.703 1.604 2.323 1.051 0.126 0.827 -3.996

0.039 0 0 0.109 0.02 0.293 0.9 0.408 0

0.4927587 0.2033686 0.0843447 0.9285573 1.060054 0.7936582 0.6914863 0.9613751 0.5341982

0.9818403 0.4952185 0.4671351 2.101885 1.989602 2.149544 1.521265 1.101707 0.8070925

[95% con£. interval]

ModelB bef 1980 1980-89 1990-97 blacks mexicans other hispanics other Model C befl980 1980-89 1990-97 blacks mexicans other hispanics other age female

78

Immigrant Workers and American Labor

work experience experience squared less than HS more thanHS

79

[95% conf. interval]

Odds ratio

Std. error

1.037087

0.0418562

0.902

0.367

0.9582112

1.122456

0.9992317 0.5526986 1.053317

0.0003928 0.1464757 0.1665284

-1.955 -2.237 0.329

0.051 0.025 0.742

0.9984622 0.3287788 0.7726542

1.000002 0.9291225 1.43593

0.7615226 0.4478394 0.2964742 1.21872 1.401794 1.043667 0.8697137 0.9617607 0.5162449 1.088059 0.9995898 0.4725562 0.9745917 0.0254489 0.0902857 0.2406038 0.0719906 0.0720755

0.1501429 0.1086695 0.1344857 0.2987035 0.2519948 0.2927806 0.196842 0.0380726 0.0646946 0.0496427 0.0004368 0.14026 0.1724209 0.0190634 0.0161818 0.0514047 0.0142057 0.011554

-1.382 -3.311 -2.68 0.807 1.879 0.152 -0.617 -0.985 -5.276 1.85 -0.939 -2.526 -0.145 -4.901 -13.417 -6.668 -13.334 -16.407

0.167 0.001 0.007 0.42 0.06 0.879 0.537 0.325 0 0.064 0.348 0.012 0.884 0 0 0 0 0

0.5174383 1.120745 0.2783402 0.7205577 0.1218618 0.7212841 0.7538379 1.97029 0.9855242 1.99389 0.6022458 1.808631 0.5581156 1.355278 0.8899612 1.039353 0.4038175 0.6599733 0.9949845 1.18984 0.9987341 1.000446 0.2641236 0.8454729 0.6890195 1.378523 0.005862 0.1104824 0.0635417 0.1282858 0.1582866 0.3657301 0.0489003 0.105984 0.052642 60.0986821

0.7376172 0.4367066 0.2862236 1.150279 1.411128 1.03212 0.8177714 0.9524951 0.5209592 1.098306 0.9995841 0.4577067 0.9693333 0.0282969 0.086731

0.1464471 0.1068892 0.1298044 0.2824509 0.2568898 0.2907066 0.1873813 0.0379817 0.0654156 0.0503242 0.0004365 0.1365863 0.1723118 0.021266 0.0156443

-1.533 -3.385 -2.758 0.57 1.892 0.112 -0.878 -1.221 -5.193 2.046 -0.953 -2.619 -0.175 -4.744 -13.555

0.125 0.001 0.006 0.569 0.059 0.911 0.38 0.222 0 0.041 0.341 0.009 0.861 0 0

0.4998421 0.2703012 0.1176738 0.7108707 0.9876628 0.5942658 0.5219039 0.8808871 0.4073057 1.003972 0.998729 0.2550213 0.6841649 0.0064869 0.0609027

z

P>lzl

ModelD befl980 1980-89 1990-97 blacks mexicans other hispanics other age female work experience exper squared less than HS more than HS agr I min I fores const/mfg

TCU

trade FIRE/ services ModelE befl980 1980-89 1990-97 blacks mexicans other hispanics other age female work experience exper squared less than HS more than HS agr /min/fores const/mfg

1.088502 0.7055561 0.6961955 1.861297 2.016155 1.792583 1.281366 1.029924 0.6663264 1.201504 1.00044 0.821482 1.373363 0.1234367 0.1235128

APPENDIX B

80

TCU trade FIRE/ services

SF

other Metro non-Metro

Odds ratio

Std. error

z

P>lzl

0.2242095 0.0698902 0.0682211 1.25093 0.7667482 0.4831946

0.0484532 0.0138633 0.0110895 0.1928156 0.117202 0.2426792

-6.919 -13.414 -16.518 1.453 -1.738 -1.448

0 0 0 0.146 0.082 0.148

[95% conf. interval]

0.1467924 0.0473777 0.0496083 0.9247634 0.5682528 0.1805575

0.3424557 0.1031001 0.0938174 1.692136 1.03458 1.29309

3

"Organize or Die": Labor's New Tactics and Immigrant Workers Rachel Sherman and Kim Voss

A

fter years of falling union density, many in the American labor movement have begun to heed the admonition, "organize or die." The AFL-CIO and some of its affiliated unions have started to put real resources and effort into organizing the unorganized, including immigrants. The approach they advocate is more aggressive, more confrontational, and more strategic than that used by most unions in the postwar period. Using an arsenal of innovative strategies, unions in a range of industries have waged high-profile organizing campaigns. Recent research suggests that these novel tactics succeed, despite fierce employer opposition and an unfavorable organizing climate. Bronfenbrenner andjuravich (1994), for example, found that union tactics accounted for more variation in the out~~:omes of NLRB representation elections than any other factor. More important, as another recent study indicates, unions that innovate in general and especially in terms of organizing are more successful in recruiting members than unions that do not (Fiorito, Jarley, and Delaney 1995). In particular, the use of an "aggressive rank and file intensive campaign" (including rank-and-file participation, highlighting issues such as dignity and justice, and devoting resources to organizing campaigns and organizer training) was associated with win rates significantly higher than those of campaigns that did not employ these tactics (Bronfenbrenner andjuravich 1998). The authors' names are in alphabetical order; each contributed equally to this collaboration. We gratefully acknowledge the financial support of the Institute of Industrial Relations at the University of California, Berkeley, as well as the helpful comments of Ruth Milkman and Sanford jacoby.

81

82

R A C H E L S H E R M A N and K I M V 0 S S

However, despite the effectiveness of these innovations, few unions are actually using many of them. In 1994, only 15 percent of the unions surveyed by Bronfenbrenner and Juravich used five or more of the new intensive tactics. Indeed, they report little change in the way most unions run organizing campaigns (1998). In this chapter we take up the issue of tactical innovation within the labor movement and its relationship to immigrant unionization. Using in-depth interviews with local union organizers and staff in northern California, as well as secondary information on particular tactics and campaigns, we look closely at the process by which new tactics are implemented in local and International unions. We argue that significant organizational innovation is required to implement new tactics, and that the impetus for such innovation typically arises from a combination of three factors: crisis within the local union, support from the International union, and the presence of innovative staff from outside the labor movement in the local. Crisis within the local union is often accompanied by a shift in employment patterns, as nonunionized immigrants replace native-born union workers.· Indeed, all the unions we studied had the potential to organize immigrant workers. The jurisdiction of the Hotel and Restaurant Employees (HERE) includes numerous immigrant workers (see Wells, this volume); some of the Service Employees International Union (SEIU) target industries, such as building services and health care, also employ high numbers of immigrants; and the jurisdiction of the United Food· and Commercial Workers (UFCW) include:;s many workplaces with immigrant workers, such as food processing firms and ethnic grocery stores. Moreover, successfully organizing immigrant workers, though different in some particulars from organizing nonimmigrant workers, is in principle the same process that leads to the successful organization of nonimmigrant workers, requiring local unions to approach workers with an understanding of their particular culture, interests, and work situation. Unions that have done a good job of organizing immigrant workers have also done well at organizing native-born workers, because of their successful organizational restructuring and adherence to a flexible model of organizing. The immigrant status of workers is certainly something organizers must keep in mind when conducting campaigns, but it is just one of several important factors, including the type of job the workers do, their race/ ethnicity, and their age. This chapter begins with a description of the innovative tactics that unions are using. We then present data on local unions in northern California, asking why some have innovated fully while others have not, and

"Organize or Die": Labor's New Tactics

83

we expose the relation of tactical to organizational innovation. Here we also discuss differences among the locals' efforts to organize immigrants. Finally, we identify factors that help to explain both tactical and organizational innovation and discuss their implications for immigrant organizing. Because it is local leaders and staff who decide to innovate and implement new tactics, variation occurs at the local level. Yet in-depth comparative analysis of particular locals is rare; most research takes as the unit of analysis either organizing campaigns (Bronfenbrenner 1993) or the International union (for an example of this as related to tactical innovation, see Delaney,Jarley, and Fio,rito 1996). We took local unions as our unit of analysis. Our research strategy was to study both locals that have adopted labor's new tactical repertoire and those that have not, so we could discover through comparison what differentiates more and less innovative locals. We began by consulting with labor leaders and labor scholars in northern California to find out which International unions active in the region had locals that were doing significant amounts of organizing, as only currently organizing locals might be using new tactics. Our informants identified three such unions: SEIU, HERE, and UFCW. We then focused on fourteen local affiliates of these International unions, first making sure that each International had both more and less innovative locals. This approach of selecting our cases by initially identifying the relevant International unions had the advantage of allowing us to compare locals both within and among Internationals, so that we could better distinguish the features common to innovative locals. Our design also reduced some potentially confounding variation: because all our Internationals organize in the same sector of the economy and in the same region of the country, sectoral and regional variation cannot account for the differences we observed between more and less innovative locals.' 1. We conducted interviews of approximately two hours with union staffers and organizers in almost all the major northern California locals affiliated with SEIU, HERE, and UFCW, a total offourteen locals. We conducted twenty-nine interviews in all, twenty-three of them with organizers and staff members. We also ir'Iterviewed five people affiliated with other labor movement institutions (the South Bay Labor Council, the AFL-CIO Organizing Department, the AFL-CIO Organizing Institute, a building trades union, and a San Francisco labor law firm). The interviewees were chosen primarily because of their role within the organization, and also because of their tenure within the local. The first round of interviews was done in early 1997; we then carried out follow-up telephone interviews with most of our original interviewees in the locals during late 1997 and early 1998. We also obtained extensive National Labor Relations Board (NLRB) data on organizing campaigns conducted by the locals we studied in the period 1985-95. In addition, we

84

R A C H E l S H E R M A N and K I M V 0 S S

Recent Tactical Innovations in the American Labor Movement

Some of the tactics unions have traditionally used to recruit new workers include organizing "hot shops" (firms where workers are enthusiastic about unionizing because of an immediate workplace grievance); focusing primarily on economic issues, especially wages and benefits; conducting top-down campaigns from union headquarters, with minimal participation by bargaining-unit members; reaching out to workers through gate leafleting, letters, and other kinds of nonpersonal contact; and dropping campaigns that did not develop quickly enough (Green and Tilly 1987; Perry 1987; Bronfenbrenner 1993). Recognition has usually been gained through the process established by the NLRB. These tactics grew increasingly ineffective over time, as employers began aggressively to manipulate the NLRB election process and to hire anti-union consulting firms on a regular basis. Major economic transformations such as the decline in manufacturing employment also helped to undermine the potency of labor's traditional tactics. For the unions we studied, the rise of "big box" retail stores and the increasingly corporate ownership of hotels and medical services were also crucial changes in the environment. Unions are now using a variety of tactics that emphasize worker participation, confrontation, pressure from arenas other than the worksite itself, and strategic planning (see Labor Research Review 17, 18, 20, and 21; Bronfenbrenner andJuravich 1994, 1998; Grabelsky and Hurd 1994; Brecher and Costello 1990;Johnston 1994; AFL-CIO 1985; Green and Tilly 1987). Here we describe four new types of tactics used in organizing campaigns: intensive worker organizing, corporate campaigns, strategic targeting, and obtaining union recognition without an NLRB election. Though not all of these tactics are new in absolute terms-many have historical precedents in 1930s labor organizing or in other social movements-they are new relative to most of the union organizing campaigns of the last forty years. The first tactic is what Bronfenbrenner andjuravich call a "rank and file intensive strategy." This includes: creating worker committees; conducting house visits, in which organizers have significant face-to-face contact with workers; focusing on issues such as justice and dignity, rather than solely on economics; and promoting solidarity actions on the job, such as wearing union buttons or organizing groups of workers to "delegate" the boss (Bronfenbrenner andjuravich 1994). This approach en-

reviewed the local and labor press, as well as International and local union publications, for information on organizing drives and contract campaigns.

"Organize or Die": Labor's New Tactics

85

courages worker militancy as well as leadership development and worker empowerment. The committee structure is especially important: it should reflect the organization of the workplace and the demographic make-up of the workforce. The focus on direct contact and worker participation is a major contrast to the organizing model of the postwar era. Another innovation is the corporate campaign (see Labor Research Review 1993; Perry 1987; Howley 1990), which includes tactics focused on expanding the arena of conflict beyond the point of production or service provision, where traditional campaigns have been centered. In corporate campaigns, unions do sophisticated research and financial analysis in order to find ways to attack the employer's image, profits, and ability to conduct business as usual; organizers describe this as "finding the Achilles' heel" of the company. For example, stockholders or potential lenders might be informed of financial problems with the company, or the union might publicize a company's regulatory violations related to consumer health and safety. In the arena of food service, for example, a "cockroach leaflet" gives stomach-turning information culled from USDA records. Unions may also attack secondary targets such as businesses that share members of the board of directors with the primary target, or subsidiary businesses of the main target. This tactic is especially effective when related companies have a higher public profile and are therefore more vulnerable to negative publicity. Sometimes unions also focus on officers or managers of the company, picketing their homes or targeting them in other jobs or public positions they hold. Third, unions use strategic targeting to decide which businesses to organize. Rather than merely organize opportunistically at "hot shops," unions target particular employers because of their strategic importance in the industry, which is determined on the basis of size, location, or other characteristics such as their competition with unionized businesses. As one organizer told us, "We're actually trying not to do hot shops because a hot shop is hot today and cold tomorrow, and also it may not make sense. We're trying to organize an industry, not one shop" (Mark, HERE Local B). 2 A fourth innovative tactic is to circumvent the NLRB electoral process, which our informants universally described as biased against unions. In the period before an election, employers often take advantage of their access to the workers to convince them to vote against the union, using a variety of coercive and I or persuasive measures. These range from hold2. To protect the confidentiality of our informants we have identified them by pseudonyms, and the locals are identified by letters rather than numbers. In a few cases, quotations are attributed to anonymous interviewees to further ensure confidentiality.

86

R A C H E L S H E RM A N and K I M V 0 S S

ing "captive audience meetings" (in which workers are shown videos of strikes and riots, or bags of groceries and automobiles that represent what their union dues could buy in a year) to raising wages and otherwise attempting to address workers' grievances. Furthermore, employers can file challenges to the bargaining unit, the workers' classifications, and other technical details, which can draw out an already lengthy process for months or years. Finally, NLRB penalties for unfair labor practices are very mild. Given these weaknesses of the NLRB process, some unions attempt to obtain voluntary recognition from employers based solely on getting 50 percent plus one of the cards signed. This process is often accompanied by a neutrality agreement, in which the employer agrees not to wage an anti-union campaign. The agreement is usually secured by using whatever external leverage the union has with the company to pressure the employer, although sometimes worker pressure is also involved. As part of the non-NLRB strategy, some unions have held elections supervised by community groups or respected citizens. In using this tactic, unions not only avoid a process they consider stacked against them but gain leverage because employers are uncomfortable with an approach that they cannot manipulate as effectively (see Howley 1990). Though card checks have long been a feature of the labor movement, and neutrality agreements have been used for many years by some unions, the combination of the two is new. Furthermore, the use of card checks in conjunction with other new tactics is "a strategy linked to action, and not just a deal with an employer. That it would be accompanied by worker actions, community actions, a whole corporate pressure campaign to push a company into neutrality, that's new" (Pamela, HERE IU). In addition, avoidance of the NLRB is often accompanied by other attempts to use labor law strategically. While organizers view the NLRB with skepticism, they will use the process to their advantage when they can, to cost the employer time and money and to generate negative publicity. Unions are also currently using some traditional tactics in novel ways. For example, though they have always tried to garner the support of politicians, unions are now making new demands of public officials such as pressuring them for neutrality agreements in city-funded development projects. Another revitalized tactic is the boycott, which, although it has a long history in the labor movement, now appears in a more aggressive incarnation: rather than simply adding the firm's name to a boycott list, the union uses intensive industry research to find ways to pressure clients to respect the boycott. Direct actions also have a long history in the labor

"Organize or Die": Labor's New Tactics

87

movement, but today unions are using civil disobedience more often, frequently with a major symbolic component designed to attract media attention. Finally, unions are creating coalitions with community groups around a variety of issues from health care to discrimination. These tactics cover a broad range, from worker mobilization to corporate pressure to political and community appeals. Organizers stress the need to use multiple tactics simultaneously. As one organizer said: I always tell new organizers, when you look at a program and look at strategy, imagine a wall full of buttons, and you push until suddenly you're successful. And you don't know which ones to push. And the ones you push this time and that bring you success, are not the same ones that you push next time to bring you success. Different things cause people to react differently'. (Scott, UFCW Local Z) Another organizer explained that "the way to think about it is, how many guns can you point at someone and blow them away" (Michelle, HERE Local B). The idea is to pressure the target from as many directions as possible, and it may be impossible to discern which particular tactic is the final straw. Our interviews and the limited literature on the subject show that though many unions use some innovative tactics, very few have adopted the entire collection of tactics described here. Only a few International unions have fully endorsed the new approach to organizing, and even their locals may continue to rely on old tactics or eschew new organizing altogether. Why, then, do some unions innovate while others do not, especially in the current, hostile climate?

Tactical Innovation and Organizational Change Tactical innovation has gone hand in hand with innovative changes in local union organization: the locals that used the highest number of innovative tactics had also undergone significant internal organizational changes, without which it would have been impossible to adopt the new tactics. We identified three types of locals. The first group, the full innovators, have all done significant organizing involving labor's new tactical repertoire and have innovated the most organizationally. The partial innovators, in contrast, have used only a few of the new tactics, and have changed little as organizations. A third group, the mandated innovators, are all currently engaged in comprehensive organizing campaigns, but

R A C H E L S H ER M A N

88

and K I M V 0 S S

they are just beginning to adopt the new repertoire and to reconstruct their organizations. In addition, their tactical and organizational innovations have come largely in response to pressures and encouragement from the International union. Full Innovators Five of the fourteen locals are full innovators: HERE locals A and B, and SEIU locals F, G, and H. Each has organized using strategic targeting, worker mobilization, non-NLRB recognition, civil disobedience, public pressure, and community alliances. The recent histories of these locals demonstrate that running m;:Uor innovative campaigns is impossible without profound organizational change. When asked about new unionization tactics, our interviewees spontaneously discussed organizational as well as tactical innovation. All the locals in this group reported having established organizing departments with full-time directors, full-time researchers, and sizable staffs of full-time organizers. All recruited bilingual organizers who mirrored the racial and ethnic background of the workers being organized. And all of them instituted new programs to train current union members to do some of the tasks involved in organizing, as well as to handle some of their own problems on the shop floor. The full innovators devote between 15. and 30 percent of their budgets to organizing, up from 2 or 3 percent before they adopted the new repertoire. To pay for organizers and researchers, these locals have either reduced the number of field representatives who handle members' grievances or required them to do substantial organizing as well. Whereas once the ratio of organizing staff to servicing staff was extremely lopsided in favor of servicing, today the ratio varies from a low of 1:2 to a high of 2:1. As a result, current members are receiving proportionately less servicing than they did in the past. These resource shifts require major changes in the roles that members play in the union. Innovative organizing tactics rely on new levels of commitment, participation, and organization on the part of members, both to support organizing and to handle some of their own worksite problems. Members have been asked to endorse aggressive organizing programs in these locals, and they usually run the local's Executive Board, which allocates funds to organizing. They are also encouraged to do more of the hard work of organizing, including going on house calls, engaging in civil disobedience, identifYing potential organizing targets, and conducting research. These new demands have often been a major challenge because of members' fears about a possible decline in servicing. However, the fully

"Organize or Die": Labor's New Tactics

89

innovating locals have tried to educate members about the importance of organizing to their own contracts and standard of living: [We] spent a lot of time with the executive board, we spend so much time with the membership .... [At] the annual leadership conference this year, the whole thing was on why we have to change, and a whole industry analysis of what's going on .... And by that I mean they look at this, and they can come to no other conclusion. Where we don't build and organize new workers, we're all going down the tubes. So that's the biggest thing, is members seeing that it's in their best interest, direct interest, that we better organize, ... So it's done by really being very logical, and thoughtful, and spending time figuring out how to present it so it's easy for folks to make a connection. (Steve, SEIU Local G) In some cases, members have defined their self-interest as new organizing, and given it priority over traditional concerns such as increasing their own wages or benefits or augmenting their strike fund. In 1996, for example, HERE Local A members voted overwhelmingly (approximately 1,600 to 200) to put the two dollars each member paid every month for a strike fund into an organizing fund instead, despite the fact that they had recently experienced a major strike. All the full-innovator locals have also begun to train members to handle more grievances by themselves, by enlisting the aid of a shop steward rather than a field representative. Such reorganizing of tasks is key in being able to shift resources to organizing. This